PSYC 2070

Psychological Statistics
Semester 1, 2017/2018

Week 2
 IBM SPSS
• IBM SPSS (Statistical Package for the Social Sciences)
• SPSS opens directly into an untitled Data Editor.

Data View Variable View

Making Graphs using SPSS
1.

2.
3.
Skewness & Kurtosis in SPSS
 Recall…Organising, summarising, &
describing data
• Frequency distributions
• Table: frequency table, grouped frequency table
• Graph: bar graph, histogram
• Shape: skewness, kurtosis

• Measures of central tendency

– Mean
– Median
– Mode
• Measures of variability / dispersion
– Range
– Variance
– Standard deviation
 Measures of central tendency
A statistical measure to determine a single score that is most
typical/ most representative of the entire dataset.

Central Tendency

Mean Median Mode

X
X Midpoint of Most
N
ranked frequently
values observed
value
 Mean
• Literal definition: Average
• Statistical definition: Sum of a
set of numbers divided by the
number of numbers in the set.
109, 90, 134, 115, 114
• Calculate the mean for this
dataset.
Mean would not work when:
• A distribution has extreme
• Population mean = mu, µ
scores (i.e., very skewed).
(myoo)
• Data are from nominal or
• Sample mean: M or X bar ordinal scale.
 Median
• Middlemost, mid-point, or most
central item in the set of
ordered numbers.
• Divides the scores so that 50%
of the scores in the distribution
have values that are equal to
or less than the median.
• No symbol.
• Calculated the same way for
populations and samples.

Original scores: X = [1,2,4,6,9,10,12,14,17]

New scores: X = [1,2,4,6,9,10,12,14, 200]
What is the median for the new scores?
 Mean vs. Median

Mean Median
• Balance point • Midpoint
• Based on • Based on middle
deviations/distance/bala location/number of
nce scores
• Change a score, Mean • Change a score, Median
will always change may not change
 Mode
• The most frequently occurring
number in a distribution (the
peak)
1,2,4,7,7,7,8,10,12,14,17
Mode: 7 because it occurs thrice

• No symbol.
• Calculated the same for
samples and populations.
• Not affected by extreme
values. Most common form of
• Measure of central tendency transport, in this data set,
for nominal data. is the bus.
When to use what
 Measures of variability / dispersion
• Describes how the scores are scattered around the central point.
• The terms variability, spread, and dispersion are synonyms.

Variability

Range Variance Standard deviation

Subtract the minimum Average of Square root of

score from the squared deviations variance
maximum score from the mean
 Measures of variability / dispersion

Quiz 1 scores
M = 7.0

Quiz 2 scores
M = 7.0
 Range
• The simplest measure of variability to calculate.
• The range = the distance / difference between the maximum
(largest) and minimum (smallest) values in a distribution.

RANGE = (Xmax– Xmin)

• Examples:
• Data set 1: [1,25,50,75,100]; Range: 100-1 = 99

• Data set 2: [48,49,50,51,52]; Range: 52-48 = 4

 Simple Deviation
• The distance of a given raw score from its mean.

Deviation score = X - μ
Example 1:
X = 53, μ = 50 therefore
Deviation score is 53 – 50 = 3
Example 2:
X = 45, μ = 50 therefore
Deviation score is 45 – 50 = -5

• The sign: tells the direction from the mean, whether the score is
located above (+) or below (-) the mean.
• The number: gives the actual distance from the mean, e.g., a
deviation score of -5 = a score that is below the mean by 5 points.
 Variance
• Simple deviation scores
must always add up to zero.

• Hence, making it rarely

used.

• So, solve the problem by

squaring the deviations.
• Variance: sum of squared
deviations divided by N, i.e., Σ(X – μ)2
number of scores/samples N
 Steps
• Find the deviation for each score
(X – μ) for each score (distance from the mean).

• Square each of these deviation

(X – μ)2 for each scores
deviation score

• Sum the squared deviations and

SS = Σ(X – μ)2 you will obtain the sum of squared
deviations (SS)

Σ(X – μ)2 • Divide the sum of squared

deviations by the number of scores
N (N)
 Steps
X μ (X – μ) (X – μ)2
(ΣX/N)

ΣX Σ(X – μ)2

Σ(X – μ)2
Variance =
N
 Standard Deviation
• Statistical definition: The square
root of the variance.
• Conceptual definition: A measure
of the average distance from the
mean.
• Tells us approximately how far
the scores vary from the mean on
average.
• A big SD implies very spread out
data.
• If the SD is small, the data is
clustered close to the mean.
 Steps
• Find the deviation for each score
(X – μ) for each score (distance from the mean).

(X – μ)2 for each • Square each of these deviation

deviation score scores

• Sum the squared deviations and

SS = Σ(X – μ)2 you will obtain the sum of squared
deviations (SS)

Σ(X – μ)2 • Divide the sum of squared

deviations by the number of scores
N (N)

• Take the square root of the variance

√Variance
Population Variance and Standard Deviation

Variance Standard Deviation

σ2 = SS / N σ = √ SS / N

OR OR

s
 ( X μ)
- 2

N
 Sample Variance and Standard Deviation

Variance Standard Deviation

s2 = SS/n - 1 = SS/df s = √ SS / n – 1 = SS
df

s 
2  -
(X M) 2
 -
( X M) 2

s
n -1 n -1
 Calculating measures of central tendency and
measures of variability using SPSS
Steps:
1. Analyse
2. Descriptive statistics
3. Frequencies
4. Move the name of the variable to Variable (s) window
5. Statistics
6. Select Mean, Median, Mode, Standard Deviation, Variance,
and Range
7. OK
1. 2.

Output
 Reporting measures of central tendency and
measures of variability according to APA style
• Three ways: narrative descriptions of the results, (2) in tables, or
(3) in graphs
• Refer: https://depts.washington.edu/psych/files/writing_center/stats.pdf

Narrative/ Text:
Ratings of essay quality for the intervention group (M = 4.3, SD =
1.2) were higher than for the control group (M = 3.3, SD = 1.0).

Table: Grade Point Average in Introduction to Psychology based on mode of credit for PSYC
1000

N Mean GPA Standard Deviation

Took class at IIUM 4,295 2.60 1.12

Took test 616 3.06 1.12

 Reporting measures of central tendency and
measures of variability according to APA style
Graph:

Reporting Median and Mode

• Mdn = Median
The median number of errors for the treatment group was 8.5,
compared to a median of 13 for the control group.
• Mode: There is no standardised notation for reporting the
mode.
 So, why need to know all these?
• Statistical procedures are designed specifically to be used with
certain types of data.
– Parametric tests: used for ratio or interval data, which falls on
a normally distributed curve.
– Non-parametric tests: for ordinal or nominal data that may or
may not fall on a normal curve.

• Using a parametric test on non-parametric data can result in

inaccurate results because of the difference in the quality of this
data.
 Exercise 1 will be distributed
today via iTaleem.
Read the instructions carefully & do
the exercise in group.
Date of Submission:
Friday, September 29th, 2017
Before 4:00 PM