Week 4

Descriptive Statistics
Course Content

Levels of Frequency
measurements distributions

Measures of central Measures of

tendency variability

Descriptive statistics
in IBM SPSS

2
 Condition or
E.g., stress level; age;
Variable characteristics that can
BASIC have different values
gender; religion
TERMINOLOGY

E.g., 0, 1, 2, 3, 4; 25, 85;

Value Number or category female / male ; Catholic /
Protestant

Score A particular person’s E.g., 4; 25; female;

value on a variable Catholic

3
 LEVEL OF MEASUREMENT

Equal interval Ratio Rank order Nominal

In addition to meeting the requirements

Equal intervals on the scale represent of an interval scale, the ratios of values The scale in which values correspond to
equal differences in the property being The scale in which the values are
along the scale should be meaningful. the relative position of things measured. categories
measured. For this to be true, the scale must have a *We still know nothing about the
true and meaningful zero point. differences between categories. We
don’t, for example, know how much
better the winner was than the
runners-up

E.g., Gender, religion

E.g., the difference between helpfulness
ratings of 1 and 2 is the same as the E.g., 2 minutes is twice as long as 1
difference between (say) 3 and 4, or 4 minutes and half as long as 4 minutes
and 5

E.g., class standing; position in a race

4
 TYPES OF VARIABLES (BASED ON LEVEL OF MEASUREMENT)

Categorical Continuous

Nominal – there are more Ordinal – same as nominal Interval – equal intervals on Ratio – The same as
Binary – there are only two
than two categories but categories have a logical the variable represent equal interval variable but the
categories
order differences in the property ratios of scores on the scale
being measured must make sense; the scale
must have a meaningful
zero
E.g., omnivore, vegetarian,
E.g., fasting or not fasting vegan, or fruitarian E.g., A, A-, B+, B, B-, …
E.g., the difference
between the score of 6 and
8 is equivalent to 13 and 15
on an intelligence test

5
 Pattern of
frequencies
over various
values

Can be described
using a frequency
table, histogram, or
frequency polygon
FREQUENCY DISTRIBUTION

Important aspect
of frequency
distribution that
we need to know –
THE SHAPE

6
 SHAPES OF FREQUENCY DISTRIBUTION

Symmetric, Unimodal Negatively-skewed distribution Positively-skewed distribution Leptokurtic distribution

distribution

Rectangular distribution Bimodal distribution Multimodal distribution Platykurtic distribution

7
The most important shape
of frequency distribution
that we need to remember
is – normal distribution
(bell-shaped curved,
symmetrical, and
unimodal). Why though? Because many naturally occurring things (such as
distribution of score for psychological variables)
have this shape. Therefore, it is expected that data
that are collected in psychological research to be
normally distributed. If the distribution of our data
is not approximately normal, we may assume that
there are errors in the data collection method
(measures / instruments and procedure) OR the
variables’ scores are naturally not normally
distributed (not all data in psychological research
should be normally distributed)

8
 Two ways a distribution can deviate from normal

Lack of symmetry (skew) Pointyness (kurtosis)

Positively or right Negatively of left Positive kurtosis - Negative kurtosis -

skewed skewed leptokurtic platykurtic

Thin in the tails and is

Frequent scores are Many scores in the tails flatter than normal
Frequent scores are and is pointy
clustered at the lower clustered at the higher
end and the tail points end and the tail points
towards the higher or towards the lower or
more positive scores more negative scores
In a normal distribution, the
values of skew and kurtosis are
0. If a distribution has values of
skew or kurtosis above or
below zero, this may indicate
deviation from normal

9
 No formula !!!

Measures of central tendency are

summary statistics that represent the centre
point or typical value of a dataset
10
 THE APPLICATION PART…

MEAN MEDIAN MODE

Should be applied to Usually applied of an Can be used for skewed
continuous data if normally ordered sample of numerical data or
distributed numerical values categorical data
Find the mean for ANXIETY Find the median for Find the mode for AGE
SCORE HOUSEHOLD INCOME

when
to
use

11
MEASURES OF VARIABILITY

• Refers to the spread of the scores within a distribution.

• Along with the central tendency, it helps in understanding the data set as
a whole
• Three major measures of variability

RANGE VARIANCE STANDARD DEVIATION

12
MEASURES OF VARIABILITY

RANGE INTERQUARTILE RANGE

Difference between the highest and Difference between the 75th and 25th
lowest score percentile

13
MEASURES OF VARIABILITY

VARIANCE STANDARD DEVIATION

The degree of spread within the A measure of how the average score
distribution deviates or spread away from the
The larger the spread, the larger the mean
variance Defined as the square root of the
variance

14
 THE APPLICATION PART…

Which one is better?

a) Normal distributed data b) Comparison large and small SD

15
MEASURES OF CENTRAL TENDENCY AND VARIABILITY
IN IBM SPSS

VARIABLE

Frequencies CATEGORICAL
Descriptive
Analyze
Statistics
Descriptives CONTINUOUS

16
Part
3
THE APPLICATION…
VARIABLE

Frequencies CATEGORICAL
Descriptive
Analyze
Statistics
Descriptives CONTINUOUS

STEPS:
1. Open your Class Exercise SPSS data file
2. Run for Option 1 – Categorical variable
3. Run for Option 2 – Continuous variable
17
 Measure of central tendency

Descriptive Measure of variability

Frequency distribution & Graph

z-score

Normal distribution/curve
Key ingredients
sample vs population

Statistics Hypothesis testing

one-sample

t-test two-sample

repeated-sample
Difference

one way
Inferential anova
two way
correlation
Relationship
regression

reliability
Psychometric
© MKS2023 factor analysis 18
Part
3
THE APPLICATION… How to do the report?
When reporting measures of central tendency and variability it is good to report the metric
of the dependent variability, if applicable. For example,
➢ The mean Beck Depression Inventory score in the drug group (M = 11.25, SD = 8.15)
was...
➢ The household income for families living in East Westernville (Median = $45,000) was...
➢ The delayed response condition (M = 560 ms; SD = 25 ms) was greater than the
immediate response condition (M = 450 ms; SD = 23 ms)...
➢ The sample as a whole was relatively young (M = 19.22, SD = 3.45).
➢ The average age of students was 19.22 years (SD = 3.45).

Extra reading…Click at the following link

https://journals.lww.com/anesthesia-analgesia/Fulltext/2017/11000/Descriptive_Statistics__Reporting_the_Answers_to.48.aspx
19
 Next class !!!
End of lesson….
Key Concepts in
Descriptive
statistics Inferential Statistics

Frequency Measures of
distribution Measures of variability
central
tendency

Range Standard
Mode deviation
Mean
Histogram
Frequency
tables
Variance
Median

20