Chapter 4

Statistics in Research
&
Processing and Data Analysis
Introduction

Types of Statistics
 Descriptive
 Inferential

1
Introduction
 The data collected is nothing more than a
group of numbers till analyzed
 Statistical analysis converts numbers into
meaningful conclusions
 What is statistics? The collecting,
summarizing, and analyzing of data
2
 Descriptive & Inferential Statistics
Descriptive Statistics Inferential Statistics

• Organize • Generalize from

samples to pops
• Summarize
• Hypothesis testing
• Simplify • Relationships
• Presentation of among variables
data

Describing data Make predictions

Descriptive Statistics
 Organize,Summarize and Simplify
Presentation of data.
Organizingdata
Summarizing data

4
Three different types of descriptive statistics:
measures of frequency, (with bar chart, line
chart, Histogram, Pie –Chart)
measures of central tendency ( mean, median,
and mode), and
measures of variability, spread or

dispersion. (range, variance, and standard deviation)

 They help displaying data visually to facilitate the
exposition of summaries of findings
5
Measures of Frequency
 indicate how often a particular phenomenon occurs.
 One of the most common ways to present frequencies is
in table format.
 frequencies may also be represented graphically in forms
such as histograms, bar graphs, or frequency polygons.
 In graphic representations, the categories are typically
plotted along the horizontal axis (x-axis), whereas the
frequencies are plotted along the vertical axis (y-axis).

6
Organizing data
Tables
Frequency Distributions
Relative Frequency Distributions

Graphs
Bar Chart or Histogram
Line Charts

7
To visualize trends in the data, it is generally useful to plot
the data even before carrying out statistical analysis

8
Summarizing Data:
Central Tendency (or Groups’ “Middle Values”)
 Mean

 Median

 Mode

Variation (or Summary of Differences Within Groups)

 Range

 Variance

 Standard Deviation

9
Measures of Central Tendency
Mean
 The sum of all the scores divided by the number of scores.
 Often referred to as the average.
 Good measure of central tendency.
 Central tendency is simply the location of the middle in a
distribution of scores.
 The mean can be misleading because it can be greatly influenced
by extreme scores (very high, or very low scores). especially if the
number of participants is small. n

x i
i 1
X
n 10
Mode
 The most common data point is called the mode.
 It may not be at the center of a distribution.
 It may give you the most likely experience rather than the
“typical” or “central” experience.
 In symmetric distributions, the mean, median, and mode
are the same.
 In skewed data, the mean and median lie further toward
the skew than the mode.

11
Mode…
 Although this measure is convenient in that it requires no
calculations, it is easily affected by chance scores,
especially if the study has a small number of participants.
 For this reason, the mode does not always give an
accurate picture of the typical behavior of the group and is
not commonly employed in many research.

12
Median
 The middle value when a variable’s values are ranked in order;
the point that divides a distribution into two equal halves.
 When data are listed in order, the median is the point at which
50% of the cases are above and 50% below it.
 The median is unaffected by outliers, making it a better measure
of central tendency, better describing the “typical person” than the
mean when data are skewed.
 If the recorded values for a variable form a symmetric distribution,
the median and mean are identical.
 In skewed data, the mean lies further toward the skew than the
median.
 commonly used with a small number of scores or when the data
contain extreme scores, known as outliers. 13
14
Measures of Dispersion
 Measures of central tendency are useful to give the
typical behavior of the group.
 But the use of measures of central tendency alone may
also obscure some important information.
  There can be cases where two groups having same
mean but, one group's scores are all close to the mean;
the other group's scores are more widely dispersed.
 We need this additional information on the dispersion, or
variability, of scores & using range , variance and
standard deviation
15
Range
 The spread, or the distance, between the lowest and
highest values of a variable.
 To get the range for a variable, you subtract its lowest
value from its highest value.
 The range, although easy to calculate, is not commonly
reported in most studies because it is sensitive to
extreme scores and thus is not always a reliable index
of variability.

16
Variance

A measure of the spread of the recorded values

on a variable. A measure of dispersion.
The larger the variance, the further the individual
cases are from the mean.
The smaller the variance, the closer the
individual scores are to the mean.

17
Standard Deviation (SD)
 A summary statistic of how much scores vary
from the mean
 Square root of the Variance
expressed in the original units of measurement
Represents the average amount of dispersion in
a sample

18
Standard Deviation (SD)…
 It is a more common way of measuring variability
 It is a number that shows how scores are spread around
the mean.
 The larger the standard deviation, the more variability
there is in a particular group of scores. Conversely, a
smaller standard deviation indicates that the group is
more homogeneous in terms of a particular behavior.
 the smaller the standard deviation, the better the mean
captures the behavior of the sample.
19
 2
n
( xi   )

i 1 n
  2
, Variance

2
n
( xi   ) Standard

i 1 n
  deviation

20
Outliers
 data that seem to be atypical of the rest of the dataset.
 The presence of outliers strongly suggests that the researcher
needs to take a careful look at the data and determine whether
the data collected from specific individuals are representative of
the data elicited from the group as a whole.
 There are times when researchers may decide not to include
outlier data in the final analysis, but if this is the case there needs
to be a principled reason for not including them beyond the fact
that they "don't fit right."
 Should researchers decide that there are principled reasons for
eliminating outlying data, a detailed explanation in the research
report needs to be provided.
21
Summary of Descriptive statistics
 Frequencies, as well as measures of central tendency,
are often presented in various studies even when they do
not relate directly to the research questions.
 because frequency measures provide a succinct
summary of the basic characteristics of the data
 allowing readers to understand the nature of the data with
minimum space expenditure.
 Providing visual representations of results in graphical
form can also contribute to a clearer understanding of any
patterns confirmed through statistical testing and can
provide an early picture of any outliers in the data. 22
Choosing the right test
Types of Inferential Statistics

 Relationships Between Variables

 Differences Between Groups

23
Relationships Between Variables
 Linear Regression, Pearson's r:
 The general equation for the line is Y = mX + b. The
equation used in linear regression is written like this:
 Y = a + bX

 r directly measures the degree of association between two

variables ( X and Y)
 Simple linear regression computes an equation of a line which allows
researchers to predict one variable from another
24
Multiple Regression
 The procedure we will study & analyzes three
or more variables simultaneously is multiple
linear regression.

 develop “models” which relate two or more “predictor

variables” to a single predicted variable. (X1, X2 … and Y)

25
t-tests
 The t-test can be used when one wants to determine if
the means of two groups are significantly different from
one another.
 There are two types of t-tests—one is used when the
groups are independent and the other, known as a paired
t-test, is used when the groups are not independent, as
in a pretest/posttest situation when the focus is within a
group

26
Analysis of variance (ANOVA).

a statistic to measure any significant

differences between three or more
independent samples (Means)

27
28
Analysis of Covariance (ANCOVA).
 There are times when there might be a preexisting
difference among groups and the variable where
that difference is manifested is related to the
dependent variable.
 In other words, differences in means on variable X
will show up on a pretest.
 The preexisting difference will need to be controlled
for and is referred to as the covariate.
29
Multivariate Analysis of Variance (MANOVA)
 The MANOVA is part of the family of analyses of
variance.
 It differs from an ANOVA in that it has more than one
dependent variable.
 In order to appropriately use a multivariate analysis of
variance, there has to be justification for believing that the
dependent variables are related to one another.

30
Important definitions (Normal Dist. and Standard Score)
NORMAL DISTRIBUTION
 A distribution describes the clustering of
scores/behaviors.
 In a normal distribution (also known as a bell curve) the
numbers (e.g., scores on a particular test) cluster
around the midpoint.
 There is an even and decreasing distribution of scores in
both directions.
  Another characteristic of a normal distribution relates to
the standard deviation.
31
NORMAL DISTRIBUTION
 In a normal distribution, approximately 34% of the data lie within 1 standard
deviation of the mean. In other words, 34% of the data are one standard
deviation above the mean and 34% are one standard deviation below the
mean (Total of 68% data on both direction).
 Two standard deviations above and below the mean, we capture an
additional 27% for a total of 95%.
 Finally, approximately 2.13% of the data fall between 2 and 3 standard
deviations, leaving only approximately .3% of the data beyond 3 standard
deviations above and below the mean (99.7%).
 If we know that a group of scores is normally distributed and if we know
the mean and the standard deviation, we can then determine where
individuals fall within a group of scores.

32
STANDARD SCORES
 There are times when we want to compare an individual's
performance on different set-up) (22/75 and 22/50).
 One way to make a more meaningful comparison is to
convert these raw scores into standard scores.
 One of the most common standard scores are z scores
uses standard deviations to reflect the distance of a score
from a mean.
 If a score is one standard deviation above the mean, it has a
z score of +1, a score that is two standard deviations above
the mean has a z score of +2, and a score that is one
standard deviation below the mean has a z score of -1.
33