Empir Software Eng

DOI 10.1007/s10664-016-9437-5

Robust Statistical Methods for Empirical Software

Engineering

Barbara Kitchenham1 · Lech Madeyski2 · David Budgen3 ·

Jacky Keung4 · Pearl Brereton1 · Stuart Charters5 ·
Shirley Gibbs5 · Amnart Pohthong6

© The Author(s) 2016. This article is published with open access at Springerlink.com

Abstract There have been many changes in statistical theory in the past 30 years, including
increased evidence that non-robust methods may fail to detect important results. The sta-
tistical advice available to software engineering researchers needs to be updated to address
these issues. This paper aims both to explain the new results in the area of robust analysis
methods and to provide a large-scale worked example of the new methods. We summarise
the results of analyses of the Type 1 error efficiency and power of standard parametric and
non-parametric statistical tests when applied to non-normal data sets. We identify paramet-
ric and non-parametric methods that are robust to non-normality. We present an analysis of
a large-scale software engineering experiment to illustrate their use. We illustrate the use of
kernel density plots, and parametric and non-parametric methods using four different soft-
ware engineering data sets. We explain why the methods are necessary and the rationale
for selecting a specific analysis. We suggest using kernel density plots rather than box plots
to visualise data distributions. For parametric analysis, we recommend trimmed means,
which can support reliable tests of the differences between the central location of two or

Communicated by: Tony Gorschek

 Lech Madeyski
[email protected]
http://madeyski.e-informatyka.pl/

1 School of Computing and Mathematics, Keele University, Keele, Staffordshire ST5 5BG, UK
2 Faculty of Computer Science and Management, Wroclaw University of Science and Technology,
Wroclaw, Poland
3 School of Engineering & Computing Sciences, Durham University, Durham, UK
4 City University of Hong Kong, Hong Kong, People’s Republic of China
5 Department of Applied Computing, Lincoln University, Lincoln, New Zealand
6 Prince of Songkla University, Songkhla, Thailand
 Empir Software Eng

more samples. When the distribution of the data differs among groups, or we have ordinal
scale data, we recommend non-parametric methods such as Cliff’s δ or a robust rank-based
ANOVA-like method.

Keywords Empirical software engineering · Statistical methods · Robust methods ·

Robust statistical methods

1 Introduction

In 1996, the first author of this paper wrote a book on software metrics (Kitchenham 1996).
In the book chapter addressing statistical methods, her advice was to use box plots to visu-
alize data. Box plots are based on the median and fourth statistics (which are similar to
quartiles), so are more robust than any graphics based on means. If data were non-normal,
she advised the use of non-parametric methods such as Kruskal-Wallis rank tests to compare
multiple samples. With more complicated designs she advised using analysis of variance
methods (ANOVA) with transformations if necessary.
Other software engineering researchers preferred to avoid the non-parametric tests rely-
ing on the Central Limit Theorem, which proves that for any set of N identically distributed
variables, the mean of the variable values will be approximately normal, with mean, μ, and
variance, σ 2 /N . The Central Limit Theorem provides the justification for use of methods
based on the normal distribution to handle small samples, such as t-tests. Their choice was
justified by the observation that simulation studies had suggested the t-test and ANOVA
were quite robust even if some of the variances within groups differed (Box 1954).
In this paper, we discuss more recent studies of the t and F tests that show that if data
sets are not normal (that is the data sets do not originate from a Gaussian distribution), the
statistical tests may not be trustworthy. Statistical hypothesis testing can make two kinds of
error. Type I errors occur when we reject the null hypothesis when it is in fact true, which
is also called a false positive. Conventionally statisticians choose a probability level they
believe is acceptable for a Type I error, which is referred to as the α-level. It is usually set
to values of 0.05 or 0.01. Type II errors occur when we fail to reject the null hypothesis
when it is in fact false, which is also called a false negative. Statisticians usually prefer the
probability of a Type II, which is referred to as the β-level to be 0.2 or less. A related concept
is statistical power which is the probability of correctly rejecting the null hypothesis, so that
power = 1−β. Although the probability of either type of error is decreased by using larger
sample sizes, aiming for a very low α-level given a predetermined sample size will increase
the achieved β-level and reduce power. Studies of classical statistical tests under conditions
of non-normality have shown that the assumed α levels of tests are likely to be incorrect,
and the power of various tests may be unacceptably low.
In a study of 440 large-sample achievement and psychometric measures data sets,
Micceri (1989) found all to be significantly non-normal. He noted that data values were
often discrete, while distributions exhibited skewness, multiple modes, long tails, large
outlying values and contamination. In our experience, similar issues affect software engi-
neering data sets.1 The prevalence of non-normal data sets and recent studies showing poor

1 There has not been a systematic review of all publicly available software engineering data sets. However,
Whigham et al. (2015) propose the use of the logarithmic transformation for their proposed cost estimation
baseline, and suggest that non-Normality is the norm for cost estimation data sets.
Empir Software Eng

performance of classical statistical tests on such data sets, suggest that empirical software
engineers need a major re-think of the techniques used for statistical analysis. Recent statis-
tical studies have not only identified analysis problems, they have also introduced methods
of addressing these problems. In this paper we identify a number of robust methods that
address the problems associated with non-normal data.2
Interest in robust methods dates back to the early 1960’s, when Tukey and his colleagues
introduced the concept of Exploratory Data Analysis (EDA), see for example Mosteller
and Tukey (1977) or Hoaglin et al. (1983). Tukey and colleagues pointed out that although
classical statistical techniques are designed to be the best possible analysis methods when
specific assumptions apply:
“... experience and further research have forced us to recognize that classical tech-
niques can behave badly when the practical situation departs from the ideal described
by such assumptions.”
Behrens (1997) summarises EDA as involving:
– An emphasis on understanding the data using graphic representations of the data.
– A focus on tentative model building and hypothesis generation as opposed to confirma-
tory analysis.
– Use of robust measures.
– Positions of skepticism and flexibility regarding which techniques to apply.
Tukey and his colleagues introduced graphical techniques such as box plots and stem-and-
leaf displays and emphasized the importance of residual analysis. All these methods are well
known to empirical software engineering researchers. However, they were also concerned
with the construction of new types of measures (see Section 3.2 and Appendix A), which
have not been taken up by software engineering researchers.
They emphasized using robust and resistant methods that can be regarded as optimal for
a broad range of situations. They proposed the following definitions:
– Resistant measures and methods are those that provide insensitivity to localized misbe-
havior in data. Resistant methods pay attention to the main body of the data and little
to outliers.
– Robust methods are those that are insensitive to departures from assumptions related to
a specific underlying model.
In this paper we focus on robust measures and robust methods and regard resistance as being
a property of such measures and metrics.
Tukey and his colleagues preferred robust and resistant methods to non-parametric meth-
ods. They point out that distribution-free methods treat all distributions equally, but robust
and resistant methods discriminate between those that are more plausible and those that
are less plausible. To distinguish their approaches from classical methods, they introduced
new terms such as batch as an alternative to sample and fourths as opposed to quartiles.
Currently few of these terms are still in use with the exception of fourths, which are used
in the context of box plots. In this paper we will introduce methods that arose from EDA
concepts (specifically central location measures related to the median and trimmed means)
but will also emphasize the use of robust non-parametric methods as viable alternatives to

2 This part of the paper is based on a keynote paper given at the EASE-2015 conference (Kitchenham 2015).
 Empir Software Eng

parametric analysis. An important issue raised in this paper is that under certain conditions
non-parametric rank-based tests can themselves lack robustness.
We illustrate the new methods using software engineering data and analyse the results
of a large scale experiment as an example of the use of these techniques. However, before
considering the robust analysis methods, we introduce the use of kernel density plots as a
means of visualising data. These can provide more information about the distribution of a
data set than can be obtained from box plots alone.
Other researchers have started to adopt the robust statistical methods discussed in this
paper, e.g., Arcuri and Briand (2011), El-Attar (2014), Madeyski et al. (2014) and Madeyski
et al. (2012). In particular, Arcuri and Briand (2014) have undertaken an important survey
of statistical tests for use in assessing randomized algorithms in software engineering. We
agree with many of their recommendations (particularly their preference for non-parametric
methods), but, in this paper, we focus on approaches suitable for relatively small samples
such as those obtained from human-based experiments, or algorithms that give rise to rela-
tive small data sets (such as project cost estimation models), rather than the large data sets
they discuss. The main contribution of this paper is to provide an overview of the techniques
with extended examples of their use and an introduction to the underlying theory. In addi-
tion, based upon using the open source R statistical programming language (R Core Team
2015), the reproducer R package by Madeyski (2015) complements this paper, as well
as the paper by Jureczko and Madeyski (2015), with the aim of making our work repro-
ducible by others (Gandrud 2015). All of our data sets are encapsulated in the reproducer
R package we have created and made available from CRAN – the official repository of R
packages. All of the figures in the paper (except the figures in Appendices A and B, which
do not depend on data sets collected by us) are built on the fly from data sets stored in the
reproducer package.

2 Problems with conventional statistical tests

In this section we summarise the results of studies that have investigated the performance
of parametric and non-parametric statistical tests under conditions of non-normality. These
studies identify some of the problems that can occur when using conventional statistical
tests on data exhibiting characteristics found in real data sets.

2.1 Parametric tests

Student’s t distribution was intended as a small sample correction for normal data, so it
is necessary to consider what happens if the population is not normal. We consider first
the one-sample case where we want to put confidence limits on the sample mean. With a
lognormal distribution (i.e., a skewed distribution with a long tail but relatively few outliers),
(Wilcox and Keselman 2003) report that with sample size n = 20, the actual distribution
varies considerably from the t-distribution. Furthermore, the problem persists even when
n = 160. In this case, using an alpha value equal to 0.1:
– The lower tail probability of a Type I error is 0.11 rather than 0.05.
– The upper tail probability of a Type I error is 0.02 rather than 0.05.
Hence, in this case, the actual probability of a Type I error is 0.13 instead of 0.1. With
n = 200, the lower tail Type I error is 0.07 instead of 0.05.
Empir Software Eng

Wilcox and Kesleman also investigated what would happen if the distribution was
skewed and had heavy tails (i.e., a relatively large number of outliers). In this case,
with n = 20 and a normal distribution, there is a .95 probability that t will be
between −2.09 and 2.09 but the actual distribution based on 5000 samples, had 0.025
and 0.975 quantiles of −8.5 and 1.29 respectively. With n = 300, the quantiles were
−2.50 and 1.70 compared with theoretical values (under normality) of −1.96 and 1.96
respectively.
There are also problems with “contaminated” normal distributions where the majority
of the data comes from one distribution and a small percentage of the data comes from a
distribution with a much larger variance. In this case, the variance is larger than the uncon-
taminated distribution, which means that the standard deviation is relatively large and the
presence of the outliers that cause the variance inflation may be masked. Variance inflation
will also increase the likelihood of Type II errors.
In the two-sample case, if, the two groups exhibit the same amount of skewness and sam-
ple sizes are equal, the t test should perform correctly because the difference between the
mean values should be distributed symmetrically. However, empirical studies summarised
by Wilcox (2012) confirm that if distributions vary in shape, Type I errors may be incorrect.
If the variance is different in each group (i.e., the data exhibit heteroscedasticity), Ramsey
(1980) showed that the t test is robust if:
– Group sizes are equal.
– Data in each group are normally distributed.
– Sample sizes are not small, where small was defined as a sample size of n < 15 in each
group.
Box (1954) reported acceptable behaviour of the t test under heteroscedasticity (unequal
variances), but his study restricted the extent of the difference between the variances. The
maximum heteroscedasticity he studied was one variance being three times larger than the
other.
In contrast, Wilcox (1998) found problems with Type I and Type II errors with
heteroscedastic data if:
– Data were normal and sample sizes were unequal for two or more groups.
– Data were normal, sample sizes were the same and there were four or more groups.
– Data were non-normal when comparing two or more groups even if sample sizes were
equal.
Thus, recent studies imply that:
– We need large sample sizes to avoid problems with non-normal data.
– With small samples and non-normal data, t tests might be very problematic.
– Data distributions exhibiting combinations of non-normal properties usually have more
severe problems than distributions with only one non-normal property.
– Except under specific conditions, the classical parametric t and F tests are vulnerable
to non-normality and heteroscedasticity.
Overall the problem is that, although the Central Limit theory confirms that (under most
practical situations) the mean of a sample is distributed normally, there are no such guaran-
tees about the variance of a sample. With messy data sets, estimates of the variance may be
far from reliable, rendering unreliable any statistical tests, such as the t test, that rely upon
knowing the variance of a mean value.
 Empir Software Eng

2.2 Non-parametric tests

Given that there might be problems with parametric tests, what about the non-parametric
methods? Unfortunately, simulation studies have shown that the large sample approxima-
tion for the Mann-Whitney-Wilcoxon (MWW) tests and Kruskal-Wallis test are strongly
affected by unequal variances, even if sample sizes are equal. In fact they can be less robust
than the standard t test, see Zimmerman and Zumbo (1993) and Zimmerman (2000).
Furthermore, problems with the rank-methods can affect the results of statistical pack-
ages and can make the difference between finding a significant result and finding a
non-significant result. Bergmann et al. (2000) compared the results of the MWW test for
non-normal data provided by 11 different statistical packages. They note that the different
packages delivered p values ranging “from significant to non-significant at the 5 % level,
depending on whether a large-sample approximation or an exact permutation form of the
test was used and, in the former case, whether or not a correction for continuity was used
and whether or not a correction for ties was made”. They concluded that “the only accurate
form of the Wilcoxon-Mann-Whitney procedure is one in which the exact permutation null
distribution is compiled for the actual data”.
The MWW test is based on the U statistic where:
n1  
U = i=1 jn=1
2
φ xi , y j (1)
and 
  1 if xi > yj
φ xi , yj = (2)
0 if xi ≤ yj .
The Wilcoxon test is based on converting the data from two independent datasets G1 and
G2 of size n1 and n2 respectively into ranks where the ranks are based on all the data
(irrespective of which group an observation belongs to). The test statistic (W) is the sum of
ranks of observations in G1:
(n1 + 1) n1
W =U+ (3)
2
The statistical tests for W and U are both based on the assumption that there are no duplicate
values.
However, ranks have a number of specific properties, that can be seen by considering the
formulas for the sum of N integers and the sum of N squared integers:
(N + 1) N
R = iN i = (4)
2
This means the average rank is
R N +1
R= = (5)
N 2
Also
(N + 1) (2N + 1) N
iN i 2 = (6)
6
which means that the variance of N ranks is:
N (N + 1)
sR2 = (7)
6
The equations for the mean and variance of ranks make it clear that, unlike the mean and
variance of the raw variables, ranks can never converge to a finite mean and variance. As the
number of observations increase, the mean and variance of the ranks increase. Furthermore,
Empir Software Eng

if sample sizes are unequal and the null hypothesis is false (i.e., the groups differ), we
are almost certain to find large differences in the variances of each group. This variance
instability makes applying the large sample tests, which are equivalent to applying the t
test (or the F test for multiple groups) to the ranks, very unreliable. This is the reason why
the rank transform process proposed by Conover and Imam (1981) is invalid.3 In addition,
the values of U and W depend on the number of observations, so they do not lead to a
meaningful effect size.
Looking back to the definition of U , we can see that it is related to the probability that a
random observation from one group is larger than a random observation from another group.
Other more reliable non-parametric effect sizes are based on normalising U with respect to
the sample size and are discussed in Section 3.3.

3 Robust statistical methods

Firstly we consider the use of kernel density plots to visualise the distribution of data sets.
Then, we present various robust statistical methods described by Wilcox (2012), who also
provides R algorithms implementing them at his website.4

3.1 Kernel density plots

In the past, Kitchenham recommended the use of box plots to give researchers an overview
of the distribution of a data set, which could alert them to potential problems of non-
normality.5 Now, we believe that advice to be incorrect, and that kernel density plots are
often preferable. Kernel density plots are derived from smoothing histograms. Algorithms
that construct kernel density plots are available in the R language (R Core Team 2015).
Figure 1 shows a box plot and two histograms with their kernel density plots superim-
posed. The data set in each case is the same. It is a data set of development effort (man
hours) from 38 Finnish projects (Kitchenham and Känsälä 1983). The box plot and both of
the kernel density plots suggest that the data is skewed. The box plot in Fig. 1a shows the
median slightly off-centred in the box towards the origin, and has a short lower tail and a
long upper tail with a single large outlier. However, the kernel density plots in Fig. 1b, c pro-
vide more detail concerning the distribution, for example indicating the possibility that the
distribution is bi-modal and confirming that the majority of the values are relatively close
to the origin. The kernel density plots are different because the number of bins used in each
density plot is different, however, the general shape of the two functions is very similar.
This demonstrates one major advantage of kernel density plots: they are not so dependent
on bin size as histograms.
Figure 2 shows the box plot and kernel density plot of the same 38 projects after trans-
forming the data by taking logs. The box plot suggests that the transformation has improved
the distribution of the data. However, the kernel density plot suggests that the direction of
skewness has been changed and the data is still far from normal.

3 Using the rank transform process, data are converted to ranks and a standard parametric analysis is applied
to the ranked data rather than the raw data.
4 http://college.usc.edu/labs/rwilcox/home
5 There are still many circumstances when a box plot can be extremely useful, for example when comparing
a large number of related distributions.
 Empir Software Eng

(a) (b) (c)

0.00010
25000

0.00012
0.00008

0.00010
20000

0.00008
0.00006
15000

Density

Density
0.00006
0.00004
10000

0.00004
0.00002
5000

0.00002
0.00000

0.00000
0

0 10000 25000 0 10000 25000

Development Effort (hrs) Development Effort (hrs)

Fig. 1 A boxplot and two kernel density plots of the same finnish data set

Figure 3 presents results from a study of software defect prediction methods aimed
at comparing simple product based models with models including product metrics and a
process metric (Madeyski and Jureczko 2015).
It shows the box plots of the percentage of classes that need to be tested to find 80 % of
the defects using a simple product-based model and an advanced model including a process
metric. The data is based on 34 software projects (Madeyski and Jureczko 2015; Madeyski
2015). Looking at the box plots of the raw data many of us would believe it was acceptable
to use a paired t-test to determine whether the advanced algorithm was better than the simple
algorithm (that is, required fewer classes to find 80 % of defects). It is not until we view the
box plot of the difference between the raw data values in Fig. 3c that we see any indication
of the problem with this data set.
However, looking at the corresponding kernel density plots in Fig. 4, it is clear that
distributions of the two sets of observations, Fig. 4a, b, are far from normal. They have
long tails and are possibly bi-modal. Furthermore, the kernel density plot of the difference
Empir Software Eng

(a) (b)

0.4
10

0.3
9

Density

0.2
8

0.1
7

0.0
6

6 7 8 9 10 11
Development Effort (ln hrs)

Fig. 2 Box plot and kernel density plots of transformed effort data

between the paired outcome values in Fig. 4c looks even worse. Although the density close
to the origin looks fairly normal, it is clear that the data has a very long lower tail with
several extreme values. Looking back to Fig. 3c, we can see that this is a case where the
box plot provides additional useful information. Although the kernel density plot of the
difference between the paired observations in Fig. 4c seems normal close to the origin, the
corresponding box plot indicates that there are many difference observations that share the
same zero value, so the distribution is strongly non-normal at the origin.
Overall these examples suggest that the use of kernel density plots and histograms are
more likely to alert us to non-normal data than box plots, but box plots can also provide
useful additional information.

3.2 Robust parametric methods

One of the most well-known robust metrics of central location is the median. It is, however,
not ideal. Although the median is robust, it ignores all but one or two observations. This
means that estimates of the standard error of the median are not efficient. They may also be
unreliable if there are duplicate values in the data. Price and Bonett (2001) have evaluated
several estimators of the sample median and proposed a new estimator that tends to have
the smallest bias.
 Empir Software Eng

(a) (b) (c)

10
90

90

5
80

80
70

70

0
60

60

−5
50

50

−10
40

40

−15
30

30

−20

Simple Defect Prediction Model Advanced Defect Prediction Model Difference

Fig. 3 Box plots of software defect prediction data

Another common approach is to remove outliers and then use the standard mean and
variance of the remaining data. Wilcox and Keselman (2003) point out that there are two
problems with this approach:
1. Outlier detection methods based on means and standard deviations can fail to detect
outliers.
2. When extreme values are discarded, the remaining observations are no longer indepen-
dent, which invalidates the calculation of the standard error.
However, Wilcox (2012) introduces several robust measures based on removing outliers
through the use of a reliable method of detecting outliers. A related approach is called
trimming. This means removing a fixed proportion of the smallest and largest values in the
data set. These methods are explained below.
Empir Software Eng

0.030 (a)
Density

0.015
0.000

20 30 40 50 60 70 80 90
Simple Defect Prediction Model

(b)
0.000 0.010 0.020
Density

20 30 40 50 60 70 80 90
Advanced Defect Prediction Model

(c)
0.12
Density

0.06
0.00

−25 −20 −15 −10 −5 0 5 10

Difference

Fig. 4 Kernel density plots of software defect prediction data

3.2.1 Robust measures based on outlier detection

Robust outlier detection relies on a robust measure of scale such as the median of the
absolute deviations from the median (MAD), so if M is the median of a set of n observations:

MAD = median|xi − M|i=1,...,n (8)

In the case of data from a normal distribution, MAD estimates the standard deviation multi-
plied by z0.75 = 0.6745, which is the 0.75 quantile of the standard normal distribution (that
is a distribution with mean μ = 0 and variance σ = 1). Any observation from the distri-
bution has a 0.5 probability of being within plus or minus 0.6745 of the median. Therefore,
instead of MAD, analysts usually use MADN , where:

MAD
MADN = (9)
0.6745
 Empir Software Eng

MADN is preferred because, if the set of observations is normally distributed, it is an

unbiased estimate of the standard deviation. MADN can be therefore be considered a robust
measure of scale.
A value xi is then assumed to be an outlier if:
|xi − M|
>k (10)
MADN
For outlier detection, Wilcox recommends setting k to 2.24. The value 2.24 corresponds to
the 0.9875 quantile of the standard normal distribution. This criterion appears less severe
than using the theoretical upper and lower tail points of the box plot as a criterion for outlier
detection, which corresponds to z0.9965 ≈ 2.698.6 However, in practice, the upper (lower)
tail length of a box plot is decreased because the theoretical value of the upper (lower) tail
is shrunk to the nearest actual data value.
To construct a robust measure of central location Mest , k is set to 1.28:
n−i2
1.28 (MADN ) (i2 − i1 ) + i=i1 +1
x(i)
Mest = (11)
n − i 1 − i2
where x(1) , x(2) , ..., x(N) are the observations written in ascending order, i1 is the number
of points for which (xi − M)/MADN < −1.28 and i2 is the number of points for which
(xi − M)/MADN > 1.28. The value 1.28 corresponds to the 0.9 quantile of the standard
normal distribution, which means that a randomly sampled observation will have an 80 %
chance of being between plus or minus 1.28. Wilcox notes that this value is often used in the
construction of robust estimators because it guards against relatively large standard errors
but sacrifices very little data when sampling from a normal distribution.
Initially, MADN is constructed using the median of the raw data. If the estimation
process is stopped at that point Mest is referred to as the one-step M−estimator (MOS).
However, Mest can be iteratively refined by substituting the current value of Mest for the
median when calculating MADN in the next iteration. We explain the theoretical justi-
fication for Mest in Appendix A. Wilcox provides a bootstrap method for calculating the
standard error of Mest , but this must be treated with caution unless our data set is a random
sample from a defined population.
Omitting the term 1.28 (MADN ) (i2 − i1 ) and replacing the criterion for identifying an
outlier with k = 2.24, leads to another estimate called the modified one step M-estimator
(MOM). Wilcox notes that MOS is better in terms of the size of the standard error, but
MOM has advantages when using small sample sizes to test hypotheses. Wilcox provides
a bootstrap method for calculating the confidence limits of MOM but does not provide an
estimate of the standard error.

3.2.2 Trimmed and Winsorized means

Trimmed means are based on removing the X % smallest and largest values in a data set.
The optimum value of X is unknown but 20 % is a reasonable default. Wilcox suggests that
this provides a reasonable balance between achieving a small standard error and controlling
the probability of a Type 1 error. The observations in the data set that specify the values that
correspond to the bottom and top X % of observations are calculated as follows. The data

6 The theoretical value of the upper (lower) tail of the box plot equivalent is found by multiplying the box
length (which calculated as z0.75 − z.25 ) by 1.5 and adding (subtracting) it to the upper fourth (from the lower
fourth).
Empir Software Eng

needs to be sorted in ascending and given subscripts from 1 to N identifying that order. Then
the subscript of the observation corresponding to X/100 = 0.0X quantile, has the subscript:
ibottom = f loor(0.0X × N ) + 1 (12)
and the subscript of the observation corresponding to 1 − 0.0X quantile is
itop = N − ilowest + 1 (13)
where the function floor truncates the value of its parameter to the nearest integer. Then, all
observations with values lower than the value corresponding to ibottom and all observations
with values greater than itop are excluded from calculation of the trimmed mean.
Winsorized means are derived by replacing the X % lowest observations with the value
of the X % quantile and X % largest observations with the value of the (100 − X) %
quantile. This is referred to as Winsorizing the data. All observations with subscripts lower
than ibottom are replaced by the value of the observation with subscript equal to ibottom . All
observations with subscript greater than itop are replaced by the value of the observation
with the subscript itop .
Trimmed means form the basis of alternative approaches to t and F . Winsorized means,
however, are not usually used as robust central measure in their own right. They are used
as a means of obtaining the variance of trimmed means. If a data set of N data points is
Winsorized and the estimate of the variance of the Winsorized data set is sw2 calculated in the
usual way, sw2 is another robust measure of scale. Furthermore, the estimate of the variance
of trimmed mean is:
sw2
2
str = (14)
N (1 − X/100)
2 is the standard error of the trimmed mean.
The square root of str

3.2.3 Examples of robust measures of central location and spread

The goal of robust measures of central location and spread is to be resistant to “misbe-
haviour in the data”. We identify the mean as non-robust because one very large abnormal
value could make the mean value abnormally large. In contrast, the median is considered
robust because one very large abnormal value would not have any effect on the median.
This property is shared by all the other robust metrics discussed in Sections 3.2.1 and 3.2.2

Table 1 Central location and scale measures for the Effort Data with and without maximum value

Metric name Central Standard Central location Standard error

location error without maximum without maximum
(%age Change) (%age Change)

Mean 7678.2895 1157.4953 7165 (6.68 %) 1065.8918 (7.91 %)

Median 5430 1522.0595 4830 (11.05 %) 1626.3678 (6.85 %)
M-Estimator 6634.2307 1560.7222 6206.4239 (6.45 %) 1484.903 (4.86 %)
MOS 6634.2307 NA 6206.4239 (6.45 %) NA
MOM 6377.2857 NA 5658.697 (11.27 %) NA
20 % Trimmed Mean 6123.4583 1414.9294 5756.3043 (6 %) 1403.2146 (0.83 %)
20 % Winsorized Mean 6796.0263 1365.7145 6573.8649 (3.27 %) 1377.7016 (0.88 %)
 Empir Software Eng

which either remove abnormally large and abnormally small values or replace them. How-
ever, unlike the data sets used in our examples, in industry data sets are not static. They
grow as new projects are completed and existing products are updated. To investigate the
impact of data set growth, we look at how the robust metrics behave when the largest value
is removed
In Table 1 we report various measures of central location derived from the data set shown
in Fig. 1. We report the values from the full data set and from the data set after the maximum
effort value was removed.
Considering first the metrics derived from the full data set, we see that, as might be
expected in a highly skewed data set, the mean is the largest of the central value metrics
and the median is the smallest. The Mest , MOM and MOS are all derived in a similar way
and all have similar values, in fact Mest and MOS have identical values. The mean has
the smallest standard error while the standard error of the other metrics (for which standard
errors can be calculated) are similar.
Looking at the impact on the metrics after removing the maximum value from the data
set, we can see that all the values have been reduced. The median has exhibited the largest
percentage change (11 %). This might be considered unexpected because the median is
supposed to be resistant to changes at the extremes of the data set. It occurs because the
values in the data set consist of only 38 data points, which are spread over a very large
range of values (from 460 to 26670). The data points in the centre of the data set are not
close together, so when a data point is removed, it causes a large fluctuation in the median.
Originally, the median was calculated as the average of the two central values ( 5430 =
(4830 + 6030)/2), once the maximum was removed the median became the central value
of the remaining 37 values which is 4830.

(a) (b)
1.2

2.5
1.0

2.0
0.8

1.5
Density
0.6

1.0
0.4

0.5
0.2

0.0
0.0

0.0 0.4 0.8 1.2

Productivity (AKSI/Effort)

Fig. 5 COCOMO productivity data — all projects

Empir Software Eng

Table 2 Central location and spread of productivity data with and without the maximum value

Metric name Central Standard Central location Standard error

location error without maximum without maximum
(% age Change) (% age Change)

Mean 0.2725 0.0316 0.2568 (5.78 %) 0.0278 (11.96 %)

Median 0.1923 0.0387 0.192 (0.17 %) 0.0283 (26.98 %)
M-Estimator 0.2251 0.0313 0.2206 (2.03 %) 0.0298 (4.83 %)
MOS 0.2256 NA 0.2209 (2.08 %) NA
MOM 0.203 NA 0.203 (0 %) NA
20 % Trimmed Mean 0.2092 0.0291 0.2033 (2.82 %) 0.0256 (11.74 %)
20 % Winsorized Mean 0.2259 0.0284 0.212 (6.17 %) 0.0254 (10.81 %)

Of the other metrics, most exhibited a change of between 6 % and 7 %, including the
mean. The mean was not as affected by the removal of the largest value as might be expected
because there were a relatively large number of large values in the data set. In this case,
the Winsorized mean exhibited the smallest change because with 38, the observation with
itop = 31 corresponded to an observation with value 14568. Once the maximum value
was removed, the value of itop = 30 corresponded to an observation with the value 14504,
corresponding to a very small 0.4 % change in the maximum value of the Winsorized data
set. In terms of the effect of removing the maximum value on the standard error, as expected,
the standard error of the mean exhibited the largest change, and the standard error of the
trimmed mean exhibited the smallest change.
As another example, consider the original COCOMO data set (Boehm 1981). This con-
tains data on 63 software projects including staff effort measured in person hours and project
size measured in K adjusted delivered source instruction (AKDSI),7 from which we can
estimate productivity as AKDSI /Eff ort. The box plot and kernel density plot of the pro-
ductivity data are shown in Fig. 5. Both the box plot Fig. 5a and the kernel density plot
Fig. 5b agree that the data is highly skewed and contains outliers. In contrast to the Effort
data set, this data set is concentrated over a small range (0.020465 to 1.25). In addition, the
largest value is relatively far from the next smallest value 0.8833), and the central five val-
ues are very close together (0.18408, 0.1917, 0.1923, 0.1987, 0.2000). The robust measures
for the productivity data are shown in Table 2.
Given the properties of this data set it is not surprising to find that the mean exhibits a
large change when the maximum value is removed and the median exhibits only a small
change. In this case, the Winsorized mean exhibits the largest change. This is because with
the full data set, N = 63, and the value of itop was 51 corresponding to an observation
with the value 0.4333. Once the maximum value was removed, the value of itop was 50
corresponding to an observation with the value 0.3786. This corresponded to a relatively
large 12.6 % change in the maximum value of the Winsorized data set. In this case, most
of the standard errors exhibited a relatively large change with the change to the median
standard error being the largest (27.0 %).
These examples, might suggest that resistance is a somewhat relative concept in the con-
text of evolving data sets and depends on the specific nature of a data set. However, they

7 The
adjustment occurs when projects are updated rather than created as new, and is intended to reflect the
amount of new/changed lines of code needed to produce the update.
 Empir Software Eng

confirm that for skewed data with outliers, the trimmed mean will be closer to the central
point of the data set than the mean and will usually be smaller than the M − Estimator,
MOS or MOM. It will also usually have a smaller standard error than the mean, even
though the divisor (and associated degrees of freedom) will be based on N (1 − 0.0X) rather
than N .
However, the real importance of using trimmed means and other robust parametric mea-
sures is that they allow non-normal data to be analysed fairly on the raw data scale. This
is particularly important for ratio-based measures that are known to be strongly skewed,
such as productivity (effort/size) or defect rates (faults/size). In spite of the extreme non-
normality of such data, practitioners still prefer to use average productivity metrics based
on the raw data, for example, to set up baselines and identify good practice, see for example
Huijgens et al. (2013).
The problem with using the mean is that with skewed data more than 50 % of projects
have productivity values less than the mean. In the COCOMO productivity data, 62 % of
the projects had productivity values less than the mean productivity value. Using the mean
value gives an inflated value to the central location of the data set, as a result of the large
values. The median is much smaller than the mean and 49 % of the projects are less than
the median. However, since the median is only based on one or two values (depending on
whether the data set has an odd or even number of observations), it is hard to defend the
median as a trustworthy measure. In contrast to the mean, 54 % projects had productivity
values less than the trimmed mean. Furthermore, since the trimmed mean is based on 60 %
of the data set it is a more defensible estimate of the central location than the median.
The practical implication is that benchmarking initiatives that label projects with values
less than the mean as poorly-performing projects might justifiably be rejected by project
managers whose projects performed better than the median. In the case of the COCOMO
productivity data, five projects had values greater than the trimmed mean but less than the
mean. Furthermore, if the data did not include the largest value, none of the projects would
change from being classified as above the trimmed mean to below the trimmed mean.
We would also suggest that projects within plus or minus two standard errors of the
trimmed mean should be considered as exhibiting average productivity. Using this criterion,
the trimmed mean would classify projects with order statistics i = 28 to i = 39 as being
average, and there would be no change if the largest value were removed. In contrast, using
the mean and its standard error, the nine projects with order statistics i = 35 to i = 43 would
be classified as average, and if the largest value were removed, the mean would classify the
8 projects with order statistics i = 36 to i = 43 as being average. Bearing in mind that the
median value corresponds to the project with order statistic i = 32, it is clear that using the
trimmed mean identifies more projects close to the centre of the distribution as average than
does the mean.
To identify poorly and exceptionally performing projects, observations with productiv-
ity values less than the value of the observation corresponding to ibottom could be described
as poorly performing (in the COCOMO example, the observation with i = 13 which had
a value 0.07266 corresponded ibottom ). Equally, projects with productivity values greater
than the value of the observation corresponding to itop could be described as exceptionally
performing projects (in the COCOMO example the observation with i = 51 which had a
value 0.4333 corresponded to itop ). (Huijgens et al. 2013) point out the value of investigat-
ing whether poorly performing projects and exceptionally performing projects have specific
characteristics. In the case of the COCOMO productivity data, all of the poorly performing
projects were categorized as embedded projects, while the projects with the six largest pro-
ductivity values were all classified as organic projects and the remaining six exceptionally
Empir Software Eng

performing products were classified as semi-detached projects. In the next section, we fol-
low up the issue of the impact of project type on productivity in order to demonstrate how
trimming can be used to test hypotheses about non-normal data sets on the raw data scale.
Another important issue is that robust measures of spread can be generalised into robust
measures of covariance. This leads to the ability to undertake multivariate analysis and
robust regression analysis of non-normal data sets without relying on normalising transfor-
mations. Although it is beyond the scope of this paper, Wilcox (2012) discusses multivariate
methods and robust regression extensively.

3.2.4 Robust alternatives to t and F tests

The problem associated with heteroscedasticity among different samples has been known
for a long time. Welch (1938) proposed a variant of the t-test that allowed for different
variances within each group. This is the default version of the t–test in R (R Core Team
2015).
The variance of the difference between two means is calculated as:8
(n1 − 1) s12 + (n2 − 1) s22
var (x̄1 − x̄2 ) = (15)
(n1 + n2 − 2)
where x̄1 is the mean of the n1 data points in one of the groups and x̄2 is the mean of the n1
data points in the other group. This is very similar to the original t-test except that the two
variances are not combined into an overall average. The major difference between a Welch
test and a t-test is that the degrees of freedom are calculated quite differently as:
(q1 + q2 )2
df =   (16)
q1 2 q2 2
(n1 −1) + (n2 −1)

where qi = si2 /ni .

Yuen’s test uses trimmed means instead of the ordinary means together with Welch’s test
as a robust test for comparing the central location of two sets of data (Yuen 1974). This
approach can be extended to cater for repeated measures (paired) designs, multiple groups
and factorial designs. It also allows researchers to test linear combinations among mean
values. For example in a factorial experiment a researcher might want to know if three levels
of a factor are additive. For example, suppose we have a cost estimation factor such as
“Required reliability” that has three levels “Low”, “Standard” and “High”, and we believe
that this has an additive effect on productivity. If we have productivity values for projects
with the different levels of reliability, an additive hypothesis is tested using the following
linear combination of mean values:
x̂Standard − x̂Low = x̂High − x̂Standard
or equivalently, that
x̂Low + x̂High − 2x̂Standard = 0
Yuen’s method is appropriate when testing for differences between central locations, but
would not be sensitive to changes in the lower tail of a distribution of the kind that can be
seen in Fig. 4.
A disadvantage of the use of Yuen’s method is that the use of trimming and Welch’s test
means that the number of degrees of freedom are substantially reduced. This will mean we

8 This equation and the equation for the degrees of freedom are incorrect in Kitchenham (2015).
 Empir Software Eng

2.0
(a)
Density
1.0
0.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Productivity for Organic Projects

(b)
4
3
Density
2
1
0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Productivity for Semi−Detached Projects

(c)
6
Density
4
2
0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Productivity for Embedded Projects

Fig. 6 Kernel density plots for the COCOMO data set for each project type

need more observations. However, if our data are not normal, we will also need a great many
observations before we can be sure that results based on the full data set are reliable.
As an example of this approach, consider the original COCOMO data set (Boehm 1981).
As discussed in Section 3.2.3, the projects were divided into three different types (referred
to as the project mode), labelled organic, embedded and semi-detached. Using this data it is
possible to test whether the productivity of projects of each type is the same.
The histograms and kernel density plots for projects of each type are shown in Fig. 6.
Inspection of the plots confirms long tailed distributions. It also suggests that productivity is
generally highest for organic projects and lowest for embedded projects, with semi-detached
projects somewhere in between.9
Using trimmed means and the algorithms produced by Wilcox, we can test whether
there are significant differences among the trimmed means for the different modes and also

9 Comparing Figs. 5 and 6, also confirms that analysing data sets in more homogeneous subsets is likely to
make the distribution of the subsets less pathological than the distribution of the full data set.
Empir Software Eng

Table 3 COCOMO project productivity summary statistics

Project type # Projects Mean SE Trimmed mean TM SE

Organic 23 0.4368 0.0625 0.3901 0.0718

Semi-detached 12 0.291 0.0482 0.285 0.0375
Embedded 28 0.1296 0.0233 0.1052 0.0133

whether there is a linear relationship between trimmed means (Wilcox 2012). Summary
statistics of the COCOMO project productivity values are shown in Table 3.
Using Yuen’s method, an overall F -test for differences among the three groups of
projects was statistically significant (F = 18.678, df1 = 2, df2 = 14.74, p = 9.100371e−
05). Although there are 28 embedded projects, 12 semi-detached projects and 23 organic
projects, the degrees of freedom for the denominator of the F -test is 14.74 rather than the
60 that would be found in a standard analysis of variance. This is because 40 % of the data is
removed by trimming and the use of Welch’s method for unstable variances further reduces
the degrees of freedom and results in non-integer values for degrees of freedom.
Wilcox also provides an algorithm that assesses all pairwise comparisons of the trimmed
means that also adjusts the confidence intervals to allow for multiple tests. The results of this
analysis are shown in Table 4. This suggests that both organic and semi-detached projects
are more productive than embedded projects but that there is no significant difference
between semi-detached and organic project productivity.
Wilcox’s algorithm will also allow you to test linear combinations of the trimmed means,
for example, to test the hypothesis that the difference between the trimmed means is linear,
that is:

LinearCombination = T ME + T MO − 2T MSD ≈ 0 (17)

The value of the linear combination of trimmed means for the COCOMO data is −0.7706
with 95 % confidence limits (−0.2779, 0.1284). This indicates that we cannot rule out
the possibility of a linear effect. However, the degrees of freedom for this test is 18.85,
which suggests the test has a low power, which is particularly problematic if we want to be
confident that the null hypothesis is likely to be true.

3.3 Non-parametric tests

Looking at Fig. 4 rather than considering the difference between means, it might be use-
ful to ask the question “What is the probability that a random observation from the set of
simple algorithms is greater than an observation from the set of advanced algorithms”. This

Table 4 COCOMO project productivity group comparisons

Comparison TM difference Lower 95 % CL Upper 95 % CL df

E v SD −0.1798 −0.2863 −0.0733 8.8933

EvO −0.2849 −0.4664 −0.1034 14.9841
SD v O −0.1051 −0.3004 0.0902 19.6753
 Empir Software Eng

question is the rationale behind Cliff’s δ. It is also similar to the rationale for the MWW test
with the U statistic, but unlike U , δ can cope with duplicate values. First of all we need to
consider three probabilities:
p1 = P (x1i > x2i )
p2 = P (x1i = x2i )
p3 = P (x1i < x2i )
Then Cliff’s δ is defined as:
δ = p1 − p3 (18)
and is therefore the difference between the probability that a random observation from group
one is greater than a random observation from group two and the probability that a random
observation from group one is less than a random observation from group two (Cliff 1993).
This is also called the expanded success rate difference (SRD) (Kraemer and Kupfer 2006).
p1 and p2 can be used to calculate the probability of superiority (Grissom 1996):
P̂ = p1 + 0.5p2 (19)
This metric has also been called the area under the receiver curve (AUC) (Kraemer and
Kupfer 2006), the measure of stochastic superiority (Aˆ12 ) (Vargha and Delaney 2000) and
the probabilistic index (P (X > Y )) (Acion et al. 2006). Arcuri and Briand (2014) rec-
ommend using the metric for software engineering data analysis. Following Vargha and
Delaney (2000), they refer to it as the Aˆ12 metric. Also since p1 + p2 + p3 = 1:
p3 = 1 − P̂ − 0.5p2 (20)
which means
δ = 2P̂ − 1 (21)
Cliff derived the standard deviation for δ, which can be used to calculate the standard
deviation of P̂ , since varδ = 4varP̂ .
Looking at the calculation of the MWW U shown in (1), and assuming there are no
duplicate values:
U
p1 = (22)
n1 n2
and
U
p3 = 1 − (23)
n1 n2
so, that
2U
δ= −1 (24)
n1 n2
Akritas and Arnold (1994) and Brunner et al. (2002) suggested a different but related
method, which also allows for duplicate observations by using midranks. Midranks are
necessary if there are two (or more) observations with the same value, in that case, the obser-
vations are both allocated the average of the two (or more) related ranks. Their method is an
ANOVA-like method based on ranks but is robust to heteroscedasticity of group variances.
It is important because it can be used to analyse much more complicated statistical designs
than simple between-groups designs.
In the two group case, their test metric is simply the probability of superiority, P̂ . It is
calculated by first pooling all observations and calculating all Rij which are the midranks
associated with the observations xij where ij corresponds to the ith observation in group j .
The average rank for group j is:
1 nj
R̂j =  Rij (25)
nj i=1
Empir Software Eng

Then:
1  
P̂ = R̄2 − R̄1 + 0.5 (26)
n1 + n2
Wilcox (2012) reports that there is not much to choose between Cliff’s method and Brunner
et al.’s method, but that Cliff’s method may have some advantages when there are many tied
values and sample sizes are small.
The ANOVA-like method can be generalised to multiple groups, but to perform an overall
test of differences among k > 2 groups uses relative effects, where the relative effect for
group j is calculated as:  
R̂j − 0.5
q̂j = (27)
N
where N = nj . The null hypothesis in this case is that q1 = q2 = ... = qk = 0.5.
Both the ANOVA-like method and Cliff’s method can be adapted for repeated mea-
sures. This means they could be considered for analysing the software defect prediction data
shown in Fig. 4. However, since the data includes 17 duplicates, it is likely that Cliff’s δ is
more appropriate. Wilcox does not provide an implementation of the paired data test, but
the details are provided in Cliff (1993). The estimate of δ is:10
dii
dw = (28)
n
where dii = 1 if the simple software defect prediction model identified fewer classes than
the advanced model, 0 if the models identified the same number of classes, and −1 if the
simple model identified more classes. The variance of dw is:
(dii − dw )2
sd2w =  (29)
n(n − 1)
The estimated value of Cliff’s δ is −0.2647 with 95 % confidence interval
(−0.4884, −0.0410) on the assumption that the estimate is approximately normally dis-
tributed. The test value is −2.319 which has a probability of p = 0.0102. This suggests
that the predictions made by the advanced defect prediction algorithm have a significant
probability of requiring the search of fewer classes than the simple algorithm. This can be
compared with the standard Wilcoxon test which reports a p−value of 0.01577 but delivers
a warning “cannot compute exact p-values with zeroes”.
For analysing multiple repeated measures (for example, studies where many different
cost estimation algorithms are applied to many different data sets), software engineering
researchers have often adopted Friedman’s test with corresponding post-hoc tests as recom-
mended by Demšar (2006) (see, for example, Dejaeger et al. 2012). However, in a study of
the performance of Friedman’s test, Agresti and Pendergast (1986) found that for an under-
lying normal distribution, their rank transformed ANOVA test could be substantially more
powerful than the Friedman test. In a more recent paper, Tian and Wilcox (2007) compared
the Agresti-Pendergast method with the ANOVA-like method developed by Brunner and
colleagues. They found that under most conditions, the ANOVA-like method was preferable
to the Agresti-Pendergast method in terms of both Type I errors and power. The exception
occurred when there were only two repeated measures for each data set. There has been no
direct comparison of the Agresti-Pendergast and Cliff’s method for cases where there are
only two repeated measures.

10 We use d rather than δ when referring to sample-based estimates of δ.

 Empir Software Eng

3.4 Guidelines for interpreting effect size magnitude

Effect size is a name given to indicators that measure the magnitude of a treatment effect.
We agree with Arcuri and Briand (2014) that effect sizes are extremely useful, as they
provide an objective measure of the importance of the experimental effect, regardless of the
statistical significance of the test statistic. Furthermore, effect sizes are much less affected
by sample size than statistical significance and, as a result, are better indicators of practical
significance (Madeyski 2010; Urdan 2005; Stout and Ruble 1995).
Cohen (1988, 1992) was the first person to propose interpretation guidelines for effect
sizes, by suggesting criteria to define a small, a medium or a large effect for use in the
behavioural sciences. However, Cohen did not present any systematic calculation of effect
sizes from research studies as the basis for his generalizations. That is why Lipsey and
Wilson (2001) found these guidelines somewhat arbitrary, and presented different interpre-
tations of the magnitude of effect sizes based on the distribution of effect sizes for over 300
meta-analyses of psychological, behavioural, and education studies, suggesting the need for
domain specific guidelines.
To allow an interpretation of effect sizes in a software engineering context, Kampenes
et al. (2007) therefore proposed magnitude labels based on a systematic review of effect
size in 92 software engineering controlled experiments. The sample size is limited but gives
a rough estimation of what constitutes small, medium and large effect sizes in the software
engineering domain.
All these guidelines were presented in Madeyski (2010). In this paper, the guidelines
were extended to include the newest effect size indicators (Cliff’s delta and the probability
of superiority) and these are all summarised in Table 5.

Table 5 Guidelines for effect size magnitude interpretation

Effect small medium large

(Cohen 1988)
d 0.20 0.50 0.80
r 0.10 0.243 0.371
r2 0.01 0.059 0.138
(Cohen 1992)
d 0.20 0.50 0.80
r 0.10 0.30 0.50
r2 0.01 0.09 0.25
(Lipsey and Wilson 2001)
d 0.30 0.50 0.67
(Kampenes et al. 2007)
g 0.17 [0.00–0.376] 0.60 [0.378–1.000] 1.40 [1.002–3.40]
r 0.09 [0–0.193] 0.30 [0.193–0.456] 0.60 [0.456–0.868]
r2 0.008 [0–0.0372] 0.09 [0.0372–0.208] 0.36 [0.208–0.753]
Vargha and Delaney (2000) and Kraemer and Kupfer (2006)
Cliffs δ (SRD) 0.112 0.276 0.428
P S(Aˆ12 ) 0.556 0.638 0.714
Empir Software Eng

An important issue for the use of effect sizes in meta-analysis is that the variance of the
effect size needs to be estimated. Effect size variances are often quite complex to calcu-
late, but Wilcox’s software provides standard errors for the Cliff’s d and the probability of
superiority (Wilcox 2012).

4 Example derived from a multi-site experiment

This section presents a large-scale example of an analysis using robust methods. In this
section, we will demonstrate three different options for analysing our data. However, this is
for explanatory purposes only, we do not advocate trying many methods until finding one
that gives the answer you want. We return to this issue when discussing the results of the
experimental analysis.

4.1 Background to the multi-site experiment

The study described here was designed to investigate the use of multi-site studies in order to
address the problems of small sample sizes in Software Engineering experiments, see Dybå
et al. (2006) and Kampenes et al. (2007). The topic for the multi-site experiment concerned
the extent to which structured abstracts were clearer and more complete than conventional
abstracts. Specifically, the study investigated the following research question:
Are software engineering researchers likely to produce clearer and more complete
abstracts when these are written using a structured form?
A report on our experiences regarding the organisation of the multi-site experiment
(referred to using the alternative term distributed experiment) is provided elsewhere
(Budgen et al. 2013). In this paper we are only concerned with the analysis of the data that
was collected from this and used to assess the above research question.

4.2 Experimental design

Formally, our experiment set out to test the following hypotheses:

– Null Hypothesis 1: Structured and conventional abstracts written by software engineer-
ing researchers are not significantly different with respect to completeness.
– Alternative Hypothesis 1: Software engineering researchers write structured abstracts
that are significantly more complete than conventional abstracts.
– Null Hypothesis 2: Structured and conventional abstracts written by software engineer-
ing researchers are not significantly different with regard to clarity.
– Alternative Hypothesis 2: Software engineering researchers write structured abstracts
that are significantly clearer than conventional abstracts.
To address these, we asked participants to assess the clarity and completeness of abstracts
of scientific papers with an empirical element that were published by a Software Engi-
neering journal that had adopted structured abstracts, comparing them with the clarity
and completeness of both abstracts published by the same journal before it adopted struc-
tured abstracts, as well as with the abstracts published by a similar journal that did not
adopt structured abstracts. This gave us the opportunity to see whether the advantages
of structured abstracts we had observed in controlled experiments carried over into the
field.
 Empir Software Eng

4.2.1 Structure and organisation of the multi-site experiment

The abstracts were obtained from academic papers published in the Information and Soft-
ware Technology journal (IST) and the Journal of Systems and Software (JSS). These
software engineering journals are both published by Elsevier, and contain many papers with
an empirical content. The important point for our experiment was that IST began mandating
the use of structured abstracts in the time period 2009-2011 whereas JSS retained the use of
conventional abstracts.
This experiment is a quasi-experiment because we selected abstracts from particular vol-
umes of the two journals, and could not randomise the source of the structured abstracts.
Based on the categories provided by Shadish et al. (2002), the experiment can be clas-
sified as “a two-group pretest-posttest design with non-equivalent control groups”. Here
the change between pretest and posttest is provided by the transition to the use of struc-
tured abstracts over the period 2009–2011 for IST, and the non-equivalent control group is
provided by the two blocks of abstracts from JSS.
We conducted the experiment across five sites: Durham and Keele Universities (UK),
Lincoln University (New Zealand), the City University (Hong Kong), and the Prince of
Songkla University (Thailand). Subsequent to the initial experiment two further sets of data
were collected, one from students at City University (Hong Kong) and the other from stu-
dents at Wroclaw University of Science and Technology (Poland). The experiment was
organized by Budgen who prepared the experiment protocol and the experimental materials,
circulated the relevant materials to each site, and co-ordinated the responses.
An Entity-Relationship style diagram illustrating the experiment together with an
explanation of the entities and their relationships is presented in Appendix B.

4.2.2 Independent and dependent variables

For this study we can identify three independent variables:

1. The source of the abstracts (JSS; IST)
2. The time of publication (Block1; Block2) For both journals, these blocks consist of
roughly eighteen months-worth of issues within the period 2009–2011. For JSS, the
boundary between blocks was based upon date (mid-2010), whereas for IST, where
the transition from conventional to structured abstracts was gradual, with many issues
having mixed forms, the boundary is across all issues of 2010, with assignment to block
being determined by the form of the abstract.
3. The location of the study/participants (UK-2 sites, NZ, Thailand, HK, Poland)
The dependent variables for the study were measures of completeness (how well an
abstract would enable a systematic reviewer to determine the relevance of the associated
paper) and clarity (the quality of writing used). These were respectively assessed using a
set of 8 questions similar to those employed in previous studies, Budgen et al. (2011, 2008),
and a 10-point Likert-like scale. The completeness score for a specific abstract for each
judge was calculated as:
i=8
i=1 (xi )
Completenessi = (30)
QA

where xi is a numeric value for completeness question i where Y es = 1, N o = 0, P artly =

0.5, N A = 0, and QA is a count of the number of questions answered, excluding NA
Empir Software Eng

responses. Thus the completeness score for a specific abstract by a specific judge is a value
between 0 and 1. The completeness score for an abstract is:
 
AbstractCompleteness = median Completenessi (31)

where i = 1, ..., 4.

4.2.3 Participants and their roles

The participants who acted as “judges” of the abstracts were intended to be undergraduate
students studying computing in some form, and who were at approximately the same level
of technical educational attainment, approximating to two years of specialist computing
study at university, but in practice, some universities also recruited participants who were
more experienced, see Budgen et al. (2013). These were students who might be expected to
read research papers that have abstracts, but who had not yet had to write dissertations and
similar documents containing abstracts. Within the English context (Durham and Keele) this
would equate to students who were at the end of their second year of study, or beginning
their third year of study. For each site, sixteen participants were recruited locally, using
local expertise to match them to the above description. Where necessary, we paid a small
honorarium to those taking part. Participants were expected to have a reasonable level of
English, since the abstracts were in English, and so we collected data about whether or not
this was their first language. Figure 16 is a flow diagram showing a high-level overview of
the experimental process undertaken at each site.
Participants were required to act as judges for four abstracts, one taken from IST and one
from JSS in the time period prior to the introduction of structured abstracts and one from IST
and one from JSS in the time period following the adoption of structured abstracts by IST.
In addition, each abstract was evaluated by four judges. A flow diagram of the experimental
process from the viewpoint of the judges is shown in Fig. 17.

4.2.4 Experimental materials

Budgen identified all empirical papers in JSS and IST over the two defined time periods. The
number of abstracts available from each source and each time period is shown in Table 6.
Budgen then created a set of four random number sequences, based on the size of each
of the blocks of abstracts. The first four values from each sequence were used to select the
abstracts for the first site, the next four for the second site and so on until he had selected
20 abstracts from each journal and each block. In the second data collection activity (from
the universities in Hong Kong and Poland), four further abstracts were selected from each
journal and block.
All data were collected using paper forms. Budgen prepared a set of data collection forms
organized as two A5 sized pages side by side. Each of these had the abstract printed on the

Table 6 Allocation of abstracts to blocks

Id IST organisation No. JSS organisation No.

Block 1 All 2009; conventional (2010) 110 All 2009; Jan–June (2010) 132
Block 2 Structured (2010); all 2011 131 Jul–Dec (2010); all 2011 173
All IST 241 All JSS 305
 Empir Software Eng

right hand page, and the questions on the left hand page. They were also suitably coded so
that they could be tracked by the experimenter. To avoid participants guessing which abstract
was supposed to be best, Budgen removed the headings from the structured abstracts and
revised any sentences rendered ungrammatical by the removal of the headings. In addition,
the title and keywords were removed from each abstract.
The questions were derived from those used in the previous studies (Budgen et al. 2011,
2008), with modifications to address the restriction of using only those papers that had an
empirical element. For the purpose of data collection, each student judge was required to
first complete a consent form, then a short form asking for demographic information, and
would then receive the four data collection forms in the defined order,11 and one at a time.
As they completed a form it was to be returned to the experimenter, who would check that
it had been fully completed and then issue the next form. A flow diagram of the process is
shown in Fig. 15.
The details of the conduct of the experiment, and of the divergences from the plan that
occurred, are described in Budgen et al. (2013). The second data collection exercise used the
same set of 16 abstracts at two different universities: one in Hong Kong (the City University)
the other in Poland (Wroclaw University of Science and Technology).

4.3 Data analysis

The statistical design is a two-by-two factor analysis with journal as one factor and time
period as the other factor. What we are interested in is whether the the difference between
completeness of IST and JSS abstracts in the second time period is significantly greater
than the difference between completeness of IST abstracts and JSS abstracts in the first time
period. This is the interaction term in a factorial design and is sometimes referred to as a
differences in differences analysis. That is, if we had normally distributed data, we would
test whether:
x̄22 − x̄12 − (x̄21 − x̄11 ) > 0 (32)
where x̄22 is the mean of IST abstracts in period 2, x̄12 is the mean of JSS abstracts in period
2, x̄21 is the mean of IST abstracts in period 1 and , x̄11 is the mean of JSS abstracts in
period 1.
This section examines a number of approaches to analysing the data from the experiment
using robust methods.

4.3.1 Preliminary analysis

Data from each site was analysed to assess whether or not there was consensus in the assess-
ment of the abstract completeness score. The analysis was based on a oneway analysis of
variance of all abstract data collected at a specific site with “abstract” as the factor with 16
levels. We used a standard ANOVA rather than a robust equivalent for this analysis because
we wanted deviations from the mean to be emphasized and to calculate the Intra-Class Cor-
relation (I CC) (Shrout and Fleiss 1979). ICC is assessed on the same subjective scale as
the Kappa agreement statistic. The analysis is shown in Table 7 where
MSBetween − MSWithin
I CC = (33)
MSBetween

11 The order was changed for each group of judges that assessed the same abstract
Empir Software Eng

Table 7 Agreement among Judges for each site

Phase Site MSBA MSW A F p I CC

1 Keele 0.0802 0.019 4.2217 0.0001 0.7631 (Substantial)

1 Durham 0.0673 0.0253 2.6598 0.0052 0.624 (Substantial)
1 Lincoln 0.0858 0.0227 3.7767 0.0002 0.7352 (Substantial)
1 Pr. Songkla 0.0409 0.0205 1.9931 0.0362 0.4983 (Moderate)
1 Hong Kong (CU) 0.0463 0.032 1.4494 0.1636 0.31 (Fair)
2 Hong Kong (CU) 0.0429 0.0579 0.7404 0.7322 −0.3506 (Poor)
2 Wroclaw (POLAND) 0.0424 0.025 1.6955 0.0841 0.4102 (Moderate)

In Table 7, the column labelled P hase identifies whether the data collection took place
in the first phase of the experiment or the second; MSBA is the mean squares between
abstracts and MSW A is the mean squares within abstracts. It is noticeable that all the sites
where English is the first language (that is, Keele University (K), Durham University (D)
and Lincoln University (L)) achieved substantial agreement, whereas other sites did not
achieve such good agreement, although only the Hong Kong City University data in phase
2 achieved no consensus. Given the lack of consensus, we decided to omit the Hong Kong
City University phase 2 data from our subsequent analyses but to include the Hong Kong
City University phase 1 data.

(a) JSS Period 1 (b) JSS Period 2

5

5
4

4
Density

Density
3

3
2

2
1

1
0

0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0
Abstract Completeness score Abstract Completeness score

(c) IST Period 1 (d) IST Period 2

5

5
4

4
Density

Density
3

3
2

2
1

1
0

0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0
Abstract Completeness score Abstract Completeness score

Fig. 7 Analysis of the median completeness score for each abstract

 Empir Software Eng

4.3.2 Analysis of the experimental data

The main analysis is in two phases relating to the two data collection periods. In the first
phase we analyzed the data from the first 5 sites, in the second phase we used meta-analysis
to aggregate the data from both phases.

4.3.3 Phase 1 analysis

Figure 7 shows the kernel density plot of the abstract data from the original 5-sites. This
is based on the median of the four average completeness scores for each abstract with 20
abstracts per journal/time period group. We use the median since it is more robust than the
mean.
Comparing Fig. 7d with plots (a), (b) and (c), it seems that if there is any impact from the
use of the structured abstracts, it has been to reduce the likelihood of low scoring abstracts.
Looking at all four plots in Fig. 7, it appears that the spread of values as well as the central
locations differ. Overall, the distribution of the data in each of the four plots do not look
normally distributed and the change in distribution implies that the variances may not be the
same in the different groups.
These issues (change in distribution, non-normal data and possible non-stable variance)
suggest that we need to consider a robust analysis. There are three possible methods of
analysing the data:
– Trimmed mean analysis of variance testing a linear combination of the trimmed means.
– ANOVA-like rank-based analysis testing the interaction term.
– Cliff’s method adapted for differences in differences.
Using a trimmed mean factorial analysis, the mean values for completeness are shown in
Table 8 and the results are:
– The Time period effect is significant (p = 0.006)
– The Journal effect is not significant (p = 0.062)
– The Interaction effect is not significant (p = 0.065)
Testing the linear contrast directly gives an effect size of 0.1265 with 95 % confidence
limits (−0.008293 to 0.2613). The confidence interval spans zero so the effect size is not
statistically significant at the p = 0.05 level.
In the past, it has not been possible to do non-parametric rank based tests for such com-
plex designs. However, the more recent approach to rank-based ANOVA (which are also
robust to problems associated with tied values and variance heterogeneity in ranks) does
allow such an analysis (Akritas et al. 1997). Akritas et al.’s paper is extremely complicated,
but fortunately, the procedure has been automated and is available in an R procedure, see
Wilcox (2012, p. 260–261). Applying this analysis to our data gives the following results:
– The Time period effect is statistically significant with p = 0.00091.
– The Journal effect is statistically significant with p = 0.0153.

Table 8 Trimmed means for

phase 1 abstract completeness IST JSS

Period 1 0.5104 0.5097

Period 2 0.6711 0.5439
Empir Software Eng

Table 9 Relative effect sizes for

phase 1 abstract completeness IST JSS

Period 1 0.4238 0.3731

Period 2 0.7219 0.4812

– The interaction is not statistically significant with p = 0.10939.

These results indicate that the completeness of the abstracts is better for the more recent
studies and that the interaction term is not significant, which agree with the trimmed mean
analysis. However, in contrast to the trimmed mean analysis, the rank-based study suggests
that there is a significant journal effect.
As discussed in Section 3.3, when applying the ANOVA-like rank method to a design that
is more complex than a simple two group experiment, the relative effect size is calculated
as shown in Table 9.
However, the relative effect sizes do not consider the differences in differences effect
(that is, they are exactly the same values that would be obtained if the data were treated
simply as coming from a one factor experiment with four levels), so cannot act as an effect
size for meta-analysis purposes. Without an effect size and the effect size variance, we
cannot incorporate data from other independent studies using meta-analysis. For that reason
we consider another analysis approach, based on Cliff’s δ.
Using Cliff’s δ, it is straightforward to test the impact of Time period and Journal on
abstract completeness by performing two separate tests and ignoring the interaction term.
The results of this analysis show:
– The Time period effect shows Time period 2 completeness exceeds Time period 1
completeness with δ = 0.4065 and 95 % confidence interval (0.1581 to 0.6061)
– The Journal effect shows that IST completeness exceeds JSS completeness with δ =
0.2912 and 95 % confidence interval (0.02908 to 0.5159).
Cliff’s approach is not restricted to two samples. He pointed out that confidence interval
of the difference between two independent δ’s can also be assessed (see Cliff, Equation 19)
as follows:
 1
2
(d2 − d1 ) ± z α2 sd22 + sd22 (34)
This is exactly what we need for a difference in differences analysis. In this case d2 is
the difference between the probability that IST abstracts score higher than JSS abstracts in
period 2 and the probability that JSS abstracts score higher than IST abstracts in period 2
and d1 is the equivalent difference for period 1. In effect, we reduce the d-value obtained
for the period 2 observations to account for difference between the groups in period 1 (the
control situation). Using Wilcox’s algorithms to calculate the d values for each time period

Table 10 Cliff’s d for phase 1

abstract completeness Period 1 Period 2 Difference

p1 0.5075 0.735
p2 0.0425 0.02
p3 0.45 0.245
d 0.0575 0.49 0.4325
sd 0.0374 0.0278 0.0465
 Empir Software Eng

(a) JSS Period 1 (b) JSS Period 2

0.4

0.4
Density

Density
0.2

0.2
0.0

0.0
3 4 5 6 7 8 9 10 3 4 5 6 7 8 9 10
Abstract clarity score Abstract clarity score

(c) IST Period 1 (d) IST Period 2

0.4

0.4
Density

Density
0.2

0.2
0.0

0.0

3 4 5 6 7 8 9 10 3 4 5 6 7 8 9 10
Abstract clarity score Abstract clarity score

Fig. 8 Kernel density plots of the median clarity score for each abstract

and calculating the differences in differences statistics ourselves, our data gives the result
shown in Table 10.
Since z α2 = 1.96, the 95 % confidence interval for the difference of the differences is
(0.3413, 0.5237). Because the effect size is positive and the confidence interval does not
include zero, the difference in difference analysis based on Cliff’s δ suggests that the IST
abstracts are more complete than JSS abstracts after the introduction of structured abstracts,
after allowing for the fact that the IST abstracts were slightly more complete than the JSS
abstracts before the introduction of structured abstracts. This result is inconsistent with
the results found by the trimmed mean analysis and the ANOVA-like rank-based method.
However, for the purposes of this example we will continue to use Cliff’s approach.
We obtain similar results when viewing the kernel density plots of the clarity scores for
each group (see Fig. 8) and analysing the clarity data (see Table 11). In this case, the effect
size is estimated as 0.25 with 95 % confidence interval (0.1159, 0.3091). A similar result

Table 11 Ciff’s d for phase 1

abstract clarity Period 1 Period 2 Difference

p1 0.46 0.5875
p2 0.1175 0.075
p3 0.4225 0.3375
d 0.0375 0.25 0.2125
sd 0.036 0.0336 0.0493
Empir Software Eng

0.8
Abstract Type
IST1
IST2
0.7 JSS1
JSS2
Abstract Completeness

0.6
0.5
0.4
0.3
0.2

3 4 5 6 7 8
Abstract Clarity

Fig. 9 Scatter plot of abstract clarity vs completeness

is to be expected because, as shown in Fig. 9, abstract completeness and clarity are highly
correlated (Kendall’s tau = 0.518, p < 0.0001).

4.3.4 Phase 2 analysis

In this section we analyse the data from Wroclaw University of Science and Technology and
discuss how it can be aggregated with the previous data. As previously noted, the second set
of data from Hong Kong showed no evidence of consensus about abstract complexity and
clarity, so could not be used.
Analysing the data in isolation, the results for the Brunner et al. relative effect sizes for
completeness were similar to those found in the first period (see Table 12), but the effects
were not significant:
– The Time period effect is not statistically significant with p = 0.41535.
– The Journal effect is not statistically significant with p = 0.6621164.
– The interaction is not statistically significant with p = 0.44237.
Cliff’s difference of differences analysis is shown in Table 13 (completeness) and
Table 14 (clarity).
For completeness, the standard error is large enough to indicate that the effect size is not
statistically significant. Furthermore, for clarity the effect size is negative. Thus, analysed by

Table 12 Relative effect sizes

for phase 2 abstract completeness IST JSS

Period 1 0.4062 0.4609

Period 2 0.6641 0.4688
 Empir Software Eng

Table 13 Cliff’s d for phase 2

abstract completeness Period 1 Period 2 Difference

p1 0.5 0.6875
p2 0 0
p3 0.5 0.3125
d 0 0.375 0.375

itself the Polish data does not support the hypothesis that structured abstracts improve com-
pleteness and clarity. The number of abstracts is clearly insufficient to provide statistically
significant results and estimates of d have a large standard error.
The correct way to incorporate the results of data collected after the analysis of an initial
tranche of data is via meta-analysis (Braver et al. 2014). Just adding the new data to the
existing data set is wrong, since it involves deciding to collect more data after looking at the
results (John et al. 2012). Equally, meta-analysis of all six studies is not a valid approach
because the five studies in the first tranche were planned in advance (before the experiment)
as defined in the protocol. Thus, they are treated as one distributed experiment.
When undertaking a meta-analysis, it is important to decide whether to perform a fixed-
effects analysis or a random-effects analysis. Borenstein et al. (2009) discuss whether meta-
analysts should use a fixed-effect or a random-effect analysis. They suggest a fixed effects
analysis is appropriate if two conditions are met. Firstly the analysts believe that all the
studies are functionally similar, secondly the goal is to compute the common effect size
for the identical population, and not to generalise to other populations. In our case the use
of exactly the same protocol and output variables model and the limited goal of our meta-
analysis suggest that a fixed-effect size is justified.
Using the R metafor package (Viechtbauer 2010) and a fixed effects analysis, the
aggregated effect size, for the completeness data, was estimated to be 0.4315 with 95 %
confidence interval (0.3411, 0.5219).
The forest plot of the analysis is shown in Fig. 10. The meta-analysis indicates that the
second data set has no effect on the results obtained from analyzing the first data set. Braver
et al. make it clear that this is what is likely to happen with a low power replication.
For the clarity data, the effect size is reversed, and the fixed effects analysis showed
evidence of heterogeneity (Q = 4.6908, df = 1, p = 0.0303). This suggests that the
clarity data results from each data collection period should not be aggregated into an overall
effect size using a fixed-effects model.

4.4 Discussion of the multi-site example

An important issue arising from the multi-site experiment is that analyses performed using
the different non-parametric methods gave different results. We must reemphasize that we

Table 14 Ciff’s d for phase 2

abstract clarity Period 1 Period 2 Difference

p1 0.8125 0.6875
p2 0.125 0.0625
p3 0.0625 0.25
d 0.75 0.4375 −0.3125
Empir Software Eng

Study Id Effect Size with 95% CL

Study 1 0.43 [ 0.34 , 0.52 ]

Study 2 0.38 [ −0.31 , 1.06 ]

Fixed Effects Model 0.43 [ 0.34 , 0.52 ]

−0.50 0.00 0.50 1.00 1.50

Observed Outcome

Fig. 10 Forest plot of completeness results from two studies

do not advocate trying every possible method of analysis until finding one that gives a sig-
nificant result. There needs to be a good reason for rejecting or selecting a specific analysis
method.
Since the completeness and clarity metrics used in our experiment are both restricted—
completeness to between 0 and 1 and clarity to between 0 and 10, and because the kernel
density plots look as if the impact of structured abstracts is to reduce the likelihood of very
incomplete abstracts, we would expect the non-parametric analyses to be more reliable than
the trimmed mean analysis.
The contrast between the analysis using Cliff’s δ and the rank-based ANOVA-like
analysis may have occurred because the rank method:
– Uses ranks obtained across all groups, which may reduce the rank differences between
specific groups.
– Includes the mid-rank values used to cater for tied values while Cliff’s method removes
the impact of tied values.
– Uses statistical tests that allow for variance heterogeneity between groups but that result
in a reduction in the degrees of freedom for the F test.
Also, within the context of simple two group comparisons, Wilcox (2012) suggests that,
with many tied values, Cliff’s method may be a bit better than the rank-based method in
terms of achieving a Type 1 probability less than the nominal alpha level. However, he does
not discuss the impact of more complex statistical designs.
It may also be because the effect size is relatively small. Although a δ value of 0.4325
would be considered large according to Table 5, the effect is rather small compared with

Table 15 Median abstract

completeness Period 1 Period 2

IST 0.5 0.6786

JSS 0.5156 0.5781
 Empir Software Eng

that obtained in a previous controlled experiment. The median of the 20 median abstract
completeness scores for each experimental condition is shown in Table 15.
This suggests that the median score is increased by approximately 0.1 which is equivalent
to getting one additional Yes answer in the 8 completeness questions. In contrast, Budgen
et al. (2011) observed a median difference between conventional and structured abstracts
of just over 0.2 using a similar scoring method. In addition, the abstracts in the Budgen
study were all written by undergraduates who would have had little experience of writing
abstracts, whereas the abstracts in our multi-site experiment were written by the authors of
the papers. Authors, even if they were post graduate students, would be more experienced
than computer science undergraduates. Thus, the likely impact of using structured abstracts
would be greater in the previous study.
However, all of these are post-hoc justifications and we would be cautious about claim-
ing that we can reject the null hypothesis that there is no difference between structured
and conventional abstracts. It would be useful to replicate the experiment with data from a
different set of abstracts. We note that such a replication should:
– Keep the title and keywords with the abstract on the evaluation form to be more
consistent with research practice.
– Ensure that abstracts selected from JSS and IST for study should come from the same
two time periods. The observed increase on completeness between the time periods
suggests that we should have ensured that the time periods for both JSS and IST were
exactly the same. As it is, there is a risk that any conventional IST abstracts obtained
from the last six months of 2010 would have a greater completeness value than ear-
lier conventional abstracts. This would have lead to an increased average completeness
for IST period 1 abstracts, which would have reduced the likelihood of detecting a
difference in differences effect.
– Review the evaluation questions themselves to see whether they can be made more
objective. The lower levels of agreement among judges who do not have English as a
first language may be a result of the abstracts, but could also be due to problems with
the evaluation questions.
Our experimental design is not appropriate for testing hypotheses regarding overall time-
trends in abstract completeness. However, our results suggest that the overall quality of
abstracts has improved in the second time period for both JSS and IST. This could be
explained because general criticisms of abstracts in systematic literature reviews, together
with experimental results suggesting structured abstracts were likely to be more complete
than conventional abstracts, would probably have increased awareness of the need for good
quality abstracts and helped produce an overall improvement. However, to properly test the
hypothesis of a general improvement an experiment would need to test the completeness of
abstracts across a wide range of journals.
In terms of advantages of non-parametric methods, looking at Fig. 7, the multi-site exam-
ple, suggests that the new non-parametric methods are preferable to conventional analysis
methods because they are able to detect changes related to the overall distribution, not just
the mean.

5 Discussion

This section summarises arguments in favour of the use of robust statistics and identifies
limitations associated with their use.
Empir Software Eng

5.1 Arguments for the use of robust statistics

We have proposed using analysis techniques that are robust to non-normality when we have
reason to believe our data is non-normal. We have also suggested the use of Kernel density
functions to identify empirical distributions that appear non-normal. However, we have not
discussed whether we should use quantile-quantile plots (q-q plots) or statistical tests to
check for normality, nor have we discussed whether it is preferable to transform data.
With respect to q-q plots, like kernel density plots, they require the analyst to make a
judgment about whether the data is normal (or normal enough) or not. In our view, the
kernel density plots are somewhat easier to interpret, but we accept that this is a matter of
personal preference.
With respect to tests for normality, we note that advocates of robust statistical methods
usually state that tests of normality have poor power (see Wilcox 2012 or Erceg-Hurn and
Mirosevich 2008), whereas advocates of the normality tests publish papers demonstrating
that their tests achieve good power (see for example, D’Agostino et al. 1990, Mudholkar
et al. 2002, or Shapiro et al. 1968). In a more recent study, Razali and Wah (2011) compared
the Shapiro-Wilk test, the Kolmogorov-Smirnov test, the Lillefors test and the Anderson-
Darling test and concluded that the Shapiro-Wilk test is the most powerful normality test,
but that the power of all four tests was low for small sample sizes (that is, sample sizes of
30 and below). The Shapiro-Wilk test results for the data sets discussed in this paper are
shown in Table 16.
The table shows that the Shapiro-Wilk test suggests more of the data sets are normally
distributed than inspection of the Kernel density plots would indicate. In addition, if we
use normality tests and they suggest some groups have normally distributed data and some
do not, applying a transformation to all groups (which is necessary for any valid statistical
analysis) may reduce the normality of any group which had more or less normal data to
begin with. Overall, with relatively small, messy data sets it seems best to err on the side of

Table 16 Shapiro-Wilk normality test probability for example data sets (data sets available from the
reproducer R package (Madeyski 2015))

Data set Measure Data set p-value p-value for log

size of test transformed data

Finnish data Effort 38 0.0004 0.0653

Software defect prediction

Simple model % Modules 34 0.3917 0.0123
Advanced model (NDC) % Modules 34 0.0373 0.0101
Embedded Productivity 28 <0.0001 0.7734
Semi-Detached Productivity 12 0.6135 0.0161
Organic Productivity 23 0.0379 0.7103

Abstract experiment data

JSS1 Completeness 20 0.0899 0.0017
JSS2 Completeness 20 0.4853 0.8029
IST1 Completeness 20 0.3194 0.0658
IST2 Completeness 20 0.6371 0.2059
 Empir Software Eng

caution and assume that the data is non-normal. Under such circumstances adopting robust
methods may sometimes be conservative, but using non-robust analysis methods would
make the results of any analysis untrustworthy.
With respect to using transformations to make data sets more Normal, our example sug-
gests that the logarithmic transformation is not a panacea for all data sets. It is also the
case that the use of standard transformations (such as logarithms or square roots) can make
interpreting results more difficult. In addition, Wilcox and Keselman (2003) point out that:
– Simple transformations do not guard against low statistical power when dealing with
heavy-tailed distributions.
– Simple transformations can alter skewed distributions but do not deal directly with
outliers.
They recommend the use of trimmed means as an effective transformation method for heavy
tailed distributions. In this paper, we have recommended the use of trimmed means when
we are concerned about changes to the central location of a data set. In addition, trimmed
means are justified in a number of different ways:
– They are a compromise between the median (maximum trimming) and the mean (zero
trimming).
– They are a form of weighted mean.
– They are based on excluding the observations that provide least information about the
central location.
– They are in common use for scoring competitions where performance and style are
judged subjectively, for example, scoring diving competitions where the two upper and
lower values from seven assessments are discarded.
The other general analysis approach that can be used with non-Normal data is robust
non-parametric analysis. We have discussed the need for newer non-parametric tests, in par-
ticular the ANOVA-like rank-based method developed by Akristas, Brunner and colleagues,
and Cliff’s δ. The advantages of the new forms of non-parametric metrics and of tests based
on those metrics are:

– They are the best way of testing ordinal scale measures. In Software Engineering
many of our measures (other than those related to elapsed time) have no physical
basis, and are more likely to be ordinal than interval or ratio measures. For example,
function points and any measures constructed primarily from subjective assessments.
This includes metrics such as the abstract completeness score used in our example in
Section 4.
– P̂ and δ provide sensible non-parametric effect sizes. Indeed for meta-analysis,
Kromrey et al. (2005) report that Cliff’s δ outperformed Cohen’s d and Hedges g
statistics.
– For purposes of meta-analysis studies, it is possible to convert the MWW U or the
Wilcoxon W statistic into P̂ or δ. Although it should be noted that there is some dis-
agreement about terminology. For example, R reports the W statistic (that is, the sum
of the ranks of the first group) but labels it U .
– P̂ and δ do not suffer from the large scale approximation problems associated with U
or W .
– Brunner’s and Cliff’s methods are implemented in R source code provided by Wilcox.
Empir Software Eng

– Both methods can be used with more complex designs than simple between-groups
designs, including repeated measures designs. The rank-based ANOVA-like approach
can be applied to virtually all standard experimental designs, including n by m
factorials.

Furthermore, both approaches have been adopted in recent published software engineering
studies. Cliff’s δ was used in El-Attar (2014), El-Attar et al. (2012) and Tappenden and
Miller (2014). The probability of superiority metric P̂ (referred to as Aˆ12 ) was used in
Madeyski et al. (2014) and Madeyski et al. (2012).

5.2 Limitations of robust statistical methods

If data sets are normally distributed or sample sizes relatively large, the robust methods are
less powerful than the standard methods. However, in many cases, the robust methods are
designed to be reasonably powerful even if the data are normal, and they are considerably
more powerful if the data are not normally distributed or sample sizes are small.
A related issue is that all the robust methods discussed in this paper (with the exception
of Cliff’s method) will lead to a reduction in the degrees of freedom available for statistical
tests and the construction of confidence intervals. For the parametric methods, trimming
which removes large and small data values, and the use of Welch’s test both contribute to
a reduction in the degrees of freedom (compare, for example, the number of projects in
each Mode type in Table 3 with the degrees of freedom for the trimmed mean statistical test
shown in Table 4).
Finally, the use of power analysis to estimate required sample size is more complex for
robust methods. In particular, the relationship between degrees of freedom and the group
variances in Welch’s test (see (16)) complicates any power analysis for trimmed means or
the rank-based ANOVA-like method.

6 Conclusions

Classical statistical analysis methods have limitations when dealing with real data that are
skewed, and/or heavy-tailed, and/or have unstable variances. Box plots can also conceal the
extent of non-normality. We recommend using kernel density plots to inspect the distribution
of data.
Parametric tests such as t and F tests are not robust to non-normality, particularly severe
skewness and combinations of non-normal properties. For comparing the central location of
different data sets, we recommend using Yuen’s test based on trimmed means and Welch’s
test for unequal variances.
Rank-based methods such as MWW and Kruskal-Wallis have problems when statistical
tests are based on large sample approximations for the rank variance. Furthermore, since
the U and W test statistics are based on rank averages which increase as the number of
observations increase, they do not deliver reliable effect sizes. For analyses that are con-
cerned with general shifts in the distribution rather than changes in the central location or
are concerned that their data are naturally ordinal-scaled, we recommend using Cliff’s or
Brunner et al.’s methods for robust non-parametric methods with Cliff’s δ or the probability
of superiority as effect sizes.
 Empir Software Eng

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 Inter-
national License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution,
and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons license, and indicate if changes were made.

Appendix A: The theory of M-estimators

This section gives a very brief introduction to the theory of M-estimators based primarily
on Goodall (1983).
The mean of a sample can be considered as the value that minimises the sum of squared
deviations from itself. Equivalently, the variance can be considered as the function of a sam-
ple which penalises observations that differ from the mean. These ideas are shown in the
top two graphics in Fig. 11. Figure 11a shows the variance function assuming a symmet-
ric distribution about a central location t. In the context of M-estimators the function that
describes the deviations from the central value is called the objective function. The objective
function makes it clear that the expected value of the squared deviation of an observation x
from t (that is, E(x − t)2 ) gives greatest weight to values that are far from t. In other words,
the function (x − t)2 increases rapidly as x move away from t.
The derivative of the objective function is called the influencefunction. The influence
function for the mean is shown in Fig. 11b. The influence function passes through 0 at the
point t confirming that t is the value that minimises the objective function. Furthermore, it
shows that the central location is unbounded and if one value x increases towards infinity,
the mean value of the sample will likewise increase towards infinity. This is a slightly more
formal way of explaining why the mean is not a robust measure of central location.
Figure 11c, d show the objective function and influence function of the median
respectively. The objective function of the median is defined to be:
ρ (x, t) = |x − t| (35)
with the corresponding influence function:
ψ (x, t) = sign(u) (36)
⎧
⎨ +1 if u > 0
sign(u) = 0 if u = 0 (37)
⎩
−1 if u < 0
The objective function of the median in Fig. 11c can be seen to penalise values close to
the central location more heavily than the squared deviations, but does not penalise distant
points as severely. Furthermore, the influence function of the median shown in Fig. 11d con-
firms that the median is bounded and thus unaffected by a single value increasing towards
infinity.
If we regard the objective function as simply a method of weighting deviations from
the central location there is no reason to restrict ourselves to the squared deviation or the
absolute deviation. We can choose any objective function or influence function that has
desirable properties. For example, Huber’s objective function corresponds to an absolute
deviation at the extremes of the distribution and the squared deviation near the centre of the
distribution. Huber’s objective function and influence function are shown in Fig. 12.
Figure 12b shown that the influence function is linear at the centre so the central value
does not change abruptly like the median, but is constant in its tails, so the central value is
not unstable if there are a few unusually large or small values.
Empir Software Eng

Huber’s objective function is:

 1 2
2x if |x| < k
ρ= (38)
k|x| − 12 x 2 if |x| ≥ k.
with influence function: 
x if |x| < k
ψ= (39)
ksign(x) if |x| ≥ k.
Huber’s objective function can be used to estimate a central location value which is the
most efficient for a contaminated Gaussian distribution (that is, a distribution where the
majority of the distribution comes from one Gaussian distribution but a small percentage
comes from another Gaussian distribution with a much larger mean or standard deviation).
The above discussion provides the context for formally defining an M-estimate.
Definition: The M-estimate Tn (x1 , ..., xn ) for the function ρ is the value of t that
minimizes the objective function i=1n ρ(x , t).
j
The properties of ρ and ψ determine the properties of a specific M-estimator. Goodall
discusses the properties that are appropriate for a robust M-estimator of the central location:
1. The breakdown bound of the estimator should be large. The breakdown bound is the
largest possible fraction of the observations for which there is a bound on the estimate

(a) Mean Objective Function (b) Mean Influence Function

0
0

0 t 0 t
x x

(c) Median Objective Function (d) Median Influence Function

1
0
−1
0

0 t 0 t
x x

Fig. 11 The objective functions and influence functions of the mean and median
 Empir Software Eng

when that fraction is altered without restriction. For example, in the case of the mean,
as a single observation is made larger, the mean increases without bound. Therefore, the
breakdown bound of the mean is 0. For the median, the breakdown bound is 1/2 − 1/n
for even n and 1/2 − 1/(2n) for odd n. For Huber’s central location estimator, the
breakdown bound is p(k) where p(k) is the proportion of the data set greater than k. For
a trimmed mean, the breakdown bound corresponds to the extent of trimming, where a
trimming constant of γ = 0.2 causes 40 % of the sample values to be trimmed.
2. The estimate should have finite gross error sensitivity which means that the influence
function is bounded. Gross error sensitivity is the maximum effect a contaminated value
can have on the M-estimator, which corresponds to then maximum absolute value of the
influence function. For the median, Huber’s central location estimator, and the trimmed
mean, the gross error sensitivity is bounded but not for the mean.
3. The estimate should have finite local-shift sensitivity. This means that the influence
function should not have large discontinuities as can be the case for the median. The
Winsorized and trimmed means also exhibit discontinuities related to the point in the
data where the x-values are replaced either by the upper and lower γ -percentile or by 0
respectively.
4. The estimator should be resistant to very large outliers. This is one of the most important
features of a robust estimator. In terms of the influence function, it means it must have a
finite rejection point. The rejection point is the least distance from the location estimator
beyond which observations do not contribute to the value of the estimate. The mean,
median and Huber’s central location estimator do not have finite rejection points, but
the trimmed mean and Winsorised mean do.
5. Some M-estimators are maximum likelihood estimators which by definition maximise
the likelihood of getting the observed data. There is a connection between the influence
function and the underlying density of a distribution. So, if we are sure of the underly-
ing distribution, we can select an influence function that will allow us to calculate an
asymptotically efficient estimator for that distribution.
6. For Gaussian data the influence function of the mean is linear. Since given the central
limit theory, all distributions are approximately normal at their centre, we might want

(a) Huber's Objective Function

Quadratic

Linear Linear
0

−k k

(b) Huber's Influence function

−k −k

Fig. 12 Huber’s objective and influence functions

Empir Software Eng

(a) Biweight
0.3
−0.3 0.0

−1.0 −0.5 0.0 0.5 1.0

(b) Andrews
1.0
−1.0 0.0

−4 −2 0 2 4

Fig. 13 The Biweight and Andrew’s influence functions

an influence function that is linear near the centre, which means that:
ψ(x) ≈ kx (40)
where k is non-zero constant usually standardised to equal 1.
7. For a symmetric underlying distribution, the objective function of an M-estimator
should give equal weight to observations at equal distances either side of the centre.
This means that the influence function should be odd, so that:
ψ(−x) = ψ(x) (41)
Researchers have derived robust measures of central location and scale based on develop-
ing influence functions that have the desirable properties itemized above. Figure 13 shows
two such functions, the biweight influence function (also known as Tukey’s bisquare), which
has the equation: 
x(1 − x 2 )2 if |x| < 1
ψ= (42)
0 if |x| ≥ 1
and the Andrews influence function, which has the equation:

sin(x/a) if |x| ≤ aπ
ψ= (43)
0 if |x| > aπ.
Both influence functions are fairly linear about the origin and tend to zero at the extreme
values. Wilcox (2012) discusses both the use of the biweight and Huber’s ψ in the construc-
tion robust measures of scale, and how such measures are used to support robust regression.
However, these issues are beyond the scope of this paper.

Appendix B: Overview of the multi-site experiment

Figure 14 shows an E-R style diagram giving an overview of the experiment. An explanation
of the entities in the model follows.
Abstract An Abstract was selected at random from one Source. Abstracts are selected
without replacement, so an Abstract was assigned to a specific Site and no other site
 Empir Software Eng

Fig. 14 Diagram illustrating the organisation of the experiment

(with the exception that sites 6 and 7 used the same set of abstracts). Each Abstract was
assessed by 4 Judges and so has 4 associated Evaluation Forms.
Source The origin of the Abstract, being one of two journals (IST or JSS), in one of two
specific time periods.
Site A specific educational institution taking part in the experiment. In the first data col-
lection phase the sites were: Keele University (UK), Durham University (UK), PSU
(Thailand), Lincoln University (NZ), Polytechnic University (HK). In the second data

Table 17 Sequence order for

viewing abstracts for four judges Judge First Second Third Fourth
and four abstracts
J1 JSS-1 JSS-2 IST-2 IST-1
J2 JSS-2 IST-1 JSS-1 IST-2
J3 IST-1 IST-2 JSS-2 JSS-1
J4 IST-2 JSS-1 IST-1 JSS-2
Empir Software Eng

collection phase the sites were: City University (HK), Wroclaw University of Science
and Technology (Poland). Sites were each assigned 16 Abstracts.

Fig. 15 Flow diagram of the experimental material preparation process

 Empir Software Eng

Judge Each Site recruited 16 Judges intended to be students at the end of their second
year of CS/IT studies. Each Judge evaluated 4 of the 16 Abstracts assigned to his/her
Site.
Assessment Group An Assessment Group comprised 4 Judges who all assessed the
same 4 Abstracts.
Sequence Order Judges in a specific Assessment Group assessed Abstracts in one of
two balanced Sequence Orders. Each Judge in a specific Assessment Group was
intended to assess the 4 Abstracts assigned to the Assessment Group in a different
sequence (see Table 17, where JSS-1 means an abstract from JSS time period 1, JSS-2 an
abstract from JSS in time period 2, IST-1 an abstract from IST in time period 1 and IST-2
an abstract from IST in time period 2). Each Site was required to use the given sequence
for each pairing of four abstracts and four judges. Although the protocol required adopt-
ing the different sequences, in practice, some sites did not conform with the order process
(see Budgen et al. 2013).

Fig. 16 Flow diagram of the Experimental Process at each Site

Empir Software Eng

Fig. 17 Flow diagram of the experimental process from the Judge’s viewpoint

Evaluation Form Each Evaluation Form contained the completeness and clarity infor-
mation provided by a specific Judge.
Figure 15 defines the process used to select the abstracts for the experiment and construct
the experimental materials.
Figure 16 is a flow diagram showing a high-level overview of the experimental process
undertaken at each site.
Figure 17 is flow diagram of the experimental process from the viewpoint of the judges.

References

Acion L, Peterson JJ, Temple S, Arndt S (2006) Probabilistic index: an intuitive non-parametric approach to
measuring the size of treatment effects. Stat Med 25(4):591–602. doi:10.1002/sim.2256
Agresti A, Pendergast J (1986) Comparing mean ranks for repeated measures data. Communications in
Statistics - Theory and Methods 15(5):1417–1433
 Empir Software Eng

Akritas MG, Arnold SF (1994) Fully nonparametric hypotheses for factorial designs i: multivariate repeated
measures designs. J Am Stat Assoc 89(425):336–343. doi:10.1080/01621459.1994.10476475
Akritas MG, Arnold SF, Brunner E (1997) Nonparametric hypotheses and rank statistics for unbalanced
factorial designs. J Am Stat Assoc 92(437):258–265. doi:10.1080/01621459.1997.10473623
Arcuri A, Briand L (2011) A practical guide for using statistical tests to assess randomized algorithms in
software engineering. In: ACM/IEEE international conference on software engineering (ICSE), IEEE,
pp 1–10. doi:10.1145/1985793.1985795
Arcuri A, Briand L (2014) A hitchhiker’s guide to statistical tests for assessing randomized algorithms in soft-
ware engineering. Software Testing, Verification and Reliability 24(3):219–250. doi:10.1002/stvr.1486
Behrens JT (1997) Principles and procedures of exploratory data analysis. Psychol Methods 2(2):131–160
Bergmann R, Ludbrook J, Spooren WPJM (2000) Different outcomes of the Wilcoxon-Mann-Whitney test
from different statistics packages. Am Stat 54(1):72–77
Boehm BW (1981) Software engineering economics. Prentice-Hall
Borenstein M, Hedges LV, Higgins JP, Hannah RR (2009) Introduction to meta-analysis. Wiley
Box GEP (1954) Some theorems on quadratic forms applied in the study of analysis of variance prob-
lems, i. Effect of inequality of variance in the One-Way classification. Ann Math Stat 25(2):290–302.
doi:10.1214/aoms/1177728786
Braver SL, Thoemmes FJ, Rosenthal R (2014) Continuously cumulating meta-analysis and replicability.
Perspect Psychol Sci 9(3):333–342. doi:10.1177/1745691614529796
Brunner E, Munzel U, Puri ML (2002) The multivariate nonparametric Behrens–Fisher problem. Journal of
Statistical Planning and Inference 108(1–2):37–53. doi:10.1016/S0378-3758(02)00269-0
Budgen D, Kitchenham BA, Charters SM, Turner M, Brereton P, Linkman SG (2008) Presenting software
engineering results using structured abstracts: a randomised experiment. Empir Softw Eng 13(4):435–
468. doi:10.1007/s10664-008-9075-7
Budgen D, Burn AJ, Kitchenham B (2011) Reporting computing projects through structured abstracts: a
quasi-experiment. Empir Softw Eng 16(2):244–277. doi:10.1007/s10664-010-9139-3
Budgen D, Kitchenham B, Charters S, Gibbs S, Pohthong A, Keung J, Brereton P (2013) Lessons from
conducting a distributed quasi-experiment. In: 2013 ACM/IEEE international symposium on empirical
software engineering and measurement, pp 143–152. doi:10.1109/ESEM.2013.12
Cliff N (1993) Dominance statistics: ordinal analyses to answer ordinal questions. Psychol Bull 114(3):494–
509
Cohen JW (1988) Statistical power analysis for the behavioral sciences, 2nd edn. Lawrence Erlbaum
Associates, Hillsdale, New York
Cohen JW (1992) A power primer. Psychol Bull 112(1):155–159
Conover W, Imam RL (1981) Rank transformations as a bridge between parametric and nonparametric
statistics. Am Stat 35(3):124–129
D’Agostino RB, Belanger A, D’Agostino J, Ralph B (1990) A suggestion for using powerful and informative
tests of normality. Am Stat 44(4):316–321
Dejaeger K, Verbeke W, Martens D, Baesens B (2012) Data mining techniques for software effort estimation:
a comparative study. IEEE Trans Softw Eng 38(2):357–397
Demšar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30
Dybå T, Kampenes VB, Sjøberg DIK (2006) A systematic review of statistical power in software engineering
experiments. Inf Softw Technol 48(8):745–755. doi:10.1016/j.infsof.2005.08.009
El-Attar M (2014) Using SMCD to reduce inconsistencies in misuse case models: a subject-based empirical
evaluation. J Syst Softw 87:104–118. doi:10.1016/j.jss.2013.10.017
El-Attar M, Elish M, Mahmood S, Miller J (2012) Is in-depth object-oriented knowledge necessary to develop
quality robustness diagrams? Journal of Software 7(11):2538–2552. doi:10.4304/jsw.7.11.2538-2552
Erceg-Hurn DM, Mirosevich VM (2008) Modern robust statistical methods an easy way to maximize the
accuracy and power of your research. Am Psychol 63(7):591–601
Gandrud C (2015) Reproducible research with R and R studio. CRC Press
Goodall C (1983) Understanding robust and exploratory data analysis. John Wiley and Sons Inc., chap M-
Estimators of Location: An outline of the theory, pp 339–403
Grissom RJ (1996) The magical number .7 ± .2: meta-meta-analysis of the probability of superior out-
come in comparisons involving therapy, placebo, and control. J Consult Clin Psychol 64(5):973–982.
doi:10.1037/0022-006X.64.5.973
Hoaglin DC, Mosteller F, Tukey JW (eds) (1983) Understanding robust and exploratory data analysis. Wiley
Empir Software Eng

Huijgens H, van Solingen R, van Deursen A (2013) How to build a good practice software project portfolio?
Tech. Rep. TUD-SERG-2013-019, Delft University of Technology
John LK, Loewenstein G, Prelec D (2012) Measuring the prevalence of questionable research practices with
incentives for truth telling. Psychol Sci 23(5):524–532. doi:10.1177/0956797611430953
Jureczko M, Madeyski L (2015) Cross–project defect prediction with respect to code ownership model: an
empirical study. e-Informatica Software Engineering Journal 9(1):21–35. doi:10.5277/e-Inf150102
Kampenes VB, Dybå T, Hannay JE, Sjøberg DIK (2007) A systematic review of effect size in software
engineering experiments. Inf Softw Technol 49(11-12):1073–1086. doi:10.1016/j.infsof.2007.02.015
Kitchenham B (1996) Software metrics: measurement for software process improvement. Blackwell Pub-
lishers Inc.
Kitchenham B (2015) Robust statistical methods: why, what and how: keynote. In: Proceedings of the 19th
international conference on evaluation and assessment in software engineering (EASE 2015), pp 1:1–1:6.
doi:10.1145/2745802.2747956
Kitchenham B, Känsälä K (1983) Inter-item correlations among function points. In: Proceedings ICSE 15.
IEEE Computer Society Press, pp 477–480
Kraemer HC, Kupfer DJ (2006) Size of treatment effects and their importance to clinical research and
practice. Biol Psychiatry 59(11):990–996. doi:10.1016/j.biopsych.2005.09.014
Kromrey JD, Hogarty KY, Ferron JM, Hines CV, Hess MR (2005) Robustness in meta-analysis: an empir-
ical comparison of point and interval estimates of standardized mean differences and Cliff’s delta. In:
Proceedings of the joint statistical meetings, Minneapolis
Lipsey MW, Wilson DB (2001) Practical meta-analysis. Sage Publications, California
Madeyski L (2010) Test-driven development: an empirical evaluation of agile practice. Springer, Heidelberg.
doi:10.1007/978-3-642-04288-1
Madeyski L (2015) Reproducer: reproduce statistical analyses and meta-analyses. http://madeyski.
e-informatyka.pl/reproducible-research, R package (http://CRAN.R-project.org/package=reproducer)
Madeyski L, Jureczko M (2015) Which process metrics can significantly improve defect prediction models?
An empirical study. Softw Qual J 23(3):393–422. doi:10.1007/s11219-014-9241-7
Madeyski L, Orzeszyna W, Torkar R, Józala M (2012) Appendix to the paper “Overcoming the equivalent
mutant problem: a systematic literature review and a comparative experiment of second order mutation”.
http://madeyski.e-informatyka.pl/download/app/AppendixTSE.pdf
Madeyski L, Orzeszyna W, Torkar R, Józala M (2014) Overcoming the equivalent mutant problem: a sys-
tematic literature review and a comparative experiment of second order mutation. IEEE Trans Softw Eng
40(1):23–42. doi:10.1109/TSE.2013.44
Micceri T (1989) The unicorn, the normal curve, and other improbable creatures. Psychol Bull 105(1):156–
166
Mosteller F, Tukey JW (1977) Data analysis and regression: a second course in statistics. Addison-Wesley
Mudholkar GS, Marchetti CE, Lin CT (2002) Independence characterizations and testing normality against
restricted skewness–kurtosis alternatives. Journal of Statistical Planning and Inference 104(2):485–
501
Price RM, Bonett DG (2001) Estimating the variance of the sample median. J Stat Comput Simul 68(3):295–
305. doi:10.1080/00949650108812071
R Core Team (2015) R: a language and environment for statistical computing. R foundation for statistical
computing, Vienna, Austria
Ramsey PH (1980) Exact type 1 error rates for robustness of student’s t test with unequal variances. J Educ
Behav Stat 5(4):337–349. doi:10.3102/10769986005004337
Razali NM, Wah YB (2011) Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and
Anderson-Darling tests. Journal of Statistical Modeling and Analytics 2(1):21–33
Shadish WR, Cook TD, Campbell DT (2002) Experimental and quasi-experimental designs for generalized
causal inference. Houghton Mifflin, Boston
Shapiro SS, Wilk M, Chen HJ (1968) A comparative study of various tests for normality. J Am Stat Assoc
63(324):1343–1372
Shrout P, Fleiss J (1979) Intraclass correlations: uses in assessing rater reliability. Psychol Bull 86(2):420–
428. doi:10.1037/0033-2909.86.2.420
Stout DE, Ruble TL (1995) Assessing the practical significance of empirical results in accounting edu-
cation research: the use of effect size information. Journal of Accounting Education 13(3):281–
298
 Empir Software Eng

Tappenden AF, Miller J (2014) Automated cookie collection testing. ACM Trans Softw Eng Methodol
23(1):3:1–3:40. doi:10.1145/2559936
Tian T, Wilcox R (2007) A comparison of two rank tests for repeated measures designs. Journal of Modern
Applied Statistical Methods 6(1):331–335
Urdan TC (2005) Statistics in plain english, 2nd edn. Routledge, Oxon, UK
Vargha A, Delaney HD (2000) A critique and improvement of the CL common language effect size statistics
of McGraw and Wong. J Educ Behav Stat 25(2):101–132. doi:10.3102/10769986025002101
Viechtbauer W (2010) Conducting meta-analyses in R with the metafor package. J Stat Softw 36(3):1–48
Welch BL (1938) The significance of the difference between two means when the population variances are
unequal. Biometrika 29(3-4):350–362. doi:10.1093/biomet/29.3-4.350
Whigham PA, Owen C, MacDonell S (2015) A baseline model for software effort estimation. ACM Trans
Softw Eng Methodol 24(3):20:1–20:11
Wilcox RR (1998) How many discoveries have been lost by ignoring modern statistical methods? Am
Psychol 53(3):300–314
Wilcox RR (2012) Introduction to robust estimation & hypothesis testing, 3rd edn. Elsevier
Wilcox RR, Keselman HJ (2003) Modern robust data analysis methods: measures of central tendency.
Psychol Methods 8(3):254–274
Yuen KK (1974) The two-sample trimmed t for unequal population variances. Biometrika 61(1):165–170
Zimmerman DW (2000) Statistical significance levels of nonparametric tests biased by heterogeneous
variances of treatment groups. J Gen Psychol 127(4):354–364. doi:10.1080/00221300009598589
Zimmerman DW, Zumbo BD (1993) Rank transformations and the power of the Student t test and
Welch t test for non-normal populations with unequal variances. Canadian Journal of Experimental
Psychology/Revue canadienne de psychologie expérimentale 47(3):523–539

Barbara Kitchenham is Professor of Quantitative Software Engineering at Keele University in the UK. She
has worked in software engineering for over 40 years both in industry and academia. She has published over
150 software engineering journal and conference papers. Her main research interest is software measurement
and experimentation in the context of project management, quality control, risk management, and evaluation
of software technologies. Her most recent research has focused on the application of evidence-based practice
to software engineering.
Empir Software Eng

Lech Madeyski received the PhD and Habilitation (DSc) degrees in computer science from the Wroclaw
University of Technology, Poland, in 1999 and 2011 respectively. In 2014 he was a Visiting Researcher at
Keele University, UK, invited by Prof. Kitchenham, while in 2016 he was a Visiting Professor at Blekinge
Institute of Technology, Sweden. He is currently an Associate Professor at Wroclaw University of Technol-
ogy, Poland. His research focus is on empirical software engineering, software quality, mutation testing, agile
methodologies, reproducible research, search-based software engineering, and machine learning (especially
software defect prediction) in the field of software engineering. He is one of the founders of the Interna-
tional Conference on Evaluation of Novel Approaches to Software Engineering (ENASE) series. He has
published papers in prestigious journals including, e.g., IEEE Transactions on Software Engineering, Empir-
ical Software Engineering, Information and Software Technology, Software Quality Journal, IET Software,
Software Process: Improvement and Practice, Cybernetics and Systems, and a book ”Test-Driven Develop-
ment – An Empirical Evaluation of Agile Practice” published by Springer including statistical analyses and
meta-analysis of experiments. He is a member of ACM and IEEE.

David Budgen is a Professor of Software Engineering in the School of Engineering & Computing Sciences
at Durham University in the UK. His research interests include software design, design environments, health-
care computing and evidence-based software engineering. He was awarded a BSc(Hons) in Physics and a
PhD in Theoretical Physics from Durham University, following which he worked as a research scientist for
the Admiralty and then held academic positions at Stirling University and Keele University before moving
to his present post at Durham University in 2005. He is a member of the IEEE Computer Society, the ACM
and the Institution of Engineering & Technology (IET), and is a Chartered Engineer (CEng).
 Empir Software Eng

Jacky Keung received his B.Sc. (Hons) in Computer Science from the University of Sydney, and his Ph.D.
in Software Engineering from the University of New South Wales, Australia. He is Assistant Professor in
the Department of Computer Science, City University of Hong Kong. His main research area is in software
effort and cost estimation, empirical modeling and evaluation of complex systems, and intensive data mining
for software engineering datasets. He has published papers in prestigious journals including IEEE-TSE,
IEEE-SOFTWARE, EMSE and many other leading journals and conferences.

Pearl Brereton is professor of software engineering in the School of Computing and Mathematics at Keele
University in the United Kingdom. She has worked in software engineering for over 35 years researching
across a range of topics including service-oriented and component based systems and empirical software
engineering. Over recent years her research has focused on evidence-based software engineering including
the adoption and adaptation of the systematic review methodology within the software engineering domain.
She is joint author of a recently published book on Evidence-Based Software Engineering and Systematic
Reviews.
Empir Software Eng

Stuart Charters is Head of the Department of Informatics and Enabling Technologies at Lincoln University
in New Zealand. He earned his BSc(Hons) degree and PhD in Computer Science at Durham University in
the UK. His research interests in software engineering include, evidence-based software engineering, agile
practices and software visualisation.

Shirley Gibbs is a Lecturer in the Department of Informatics & Enabling Technologies at Lincoln University
New Zealand. Her research interests are at the intersection of people and computing and include workplace
computing, computing education, and the relationships between innovation, technology use, and perceptions
of ability. Her research has compared the expectations of workplace computing skill between graduates
and employers, examined the relationship between perceived and actual skill with expertise and personality
factors and investigated relationships between team members and customers using an Agile approach to
software development.
 Empir Software Eng

Amnart Pohthong is Assistant Professor of Computer Science at Prince of Songkla University in Thailand.
He earned his BSc degree in Mathematics in 1981 and MSc degree in Computer Science in 1991 from Prince
of Songkla University. He received a PhD in Computer Science in 2000 from Keele University in the UK.
He has been working at Prince of Songkla University since 1983. Over the years he has held several posi-
tions and experienced in teaching, research, and administrative work. His research interests include software
engineering, computer-based learning, and knowledge management. He has published several journal and
conference papers.