Part 2
Dr. Christine Pereira Academic Skills Adviser ask@[Link]
Choose appropriate numerical summaries Choose appropriate graphical summaries Charts and graphs in SPSS Descriptive statistics in SPSS
Frequency tables in SPSS
Located
on Blackboard in the following:
Academic Skills (ASK) section Workshop slides Statistics and SPSS workshops Basics of SPSS Workshop Select [Link]
Part 1:
Representing Data Graphically
All variables listed here
Build your chart here
Select chart type
Select chart format
One
Nominal or Ordinal Variable
Bar Chart
Pie Chart
What is your ethnicity?
(please tick one)
White/European Asian West Indian African Other
Produce a bar chart and a pie chart for Ethnicity.
Bar Charts
Simple Bar Chart
Pie Chart
Pie Chart
Two
Nominal or Ordinal Variables
Clustered Bar Chart
Two
Nominal or Ordinal Variables
Stacked Bar Chart
I am very committed to this organisation.
Strongly Disagree 1 Disagree Neutral Agree
Strongly Agree 5
Gender
(please tick one)
Male
Female
Produce a clustered bar chart or a stacked bar chart to summarise commitment by gender.
Bar Charts
Clustered Bar Chart Stacked Bar Chart
One
Scale Variable
Histogram
Boxplot
What is your gross annual income?
Produce a histogram and a boxplot to summarise gross annual income.
Histograms
Simple Histogram
Boxplots
1D Boxplot
Two
Scale Variables
Scatterplot
Scatterplots
Simple Scatterplot
What is your gross annual income?
Age last birthday (in years)?
Produce a scatterplot to show the relationship between income and age.
Two or more
Related Scale
Variables
Related refers to Repeated Measures
Bar Chart of Means
Line Chart of Means
Produce a bar/line chart of means to summarise employees total competency score at each of the 3 time periods.
Bar Charts
Simple Bar Chart
Line Charts
Simple Line Chart
One Scale & One
Categorical
Variable
Bar Chart of Means
Line Chart of Means
Rated Skill of work:
Unskilled 1 Semi-Skilled 2 Fairly Skilled 3 Highly Skilled 4
What is your gross annual income? Produce a bar/line chart of means to summarise income by rated skill.
Bar Charts
Simple Bar Chart
Line Charts
Simple Line Chart
Right Click
Edit Content In Separate Window
Reduce chart size and Apply changes
Right Click
to view more options
E.g., Transpose Chart
Theres
much more in Chart Editor You can change:
Colours
Fonts Axis
labels Data labels Insert trend lines Rearrange (or remove) variables the list goes on
For
more detailed examples go to:
Academic
Skills on Blackboard Statistics and SPSS Statistics using SPSS: Resources Charts and Graphs Find the slides entitled: SPSS Chart Builder
Part 2:
Representing Data Numerically
(All Descriptive Stats)
(Limited Descriptive Stats)
(Limited Descriptive Stats)
Mean, Median, Mode
Median = The middle value in an ordered data set. It splits the data into 2 equal parts. Mode = Most frequent value.
(All Descriptive Stats)
(Limited Descriptive Stats)
(Limited Descriptive Stats)
Mean, Median, Mode Stdev, Variance Range, IQR Min, Max
Standard Deviation & Variance Measure how spread out the data is with respect to the mean
Range Spread of the data (Max Min)
(All Descriptive Stats)
(Limited Descriptive Stats)
(Limited Descriptive Stats)
Mean, Median, Mode Stdev, Variance Range, IQR Min, Max Skewness, Kurtosis Skewness & Kurtosis Measure how close a histogram is to a normal distribution.
(All Descriptive Stats)
(Limited Descriptive Stats)
(Limited Descriptive Stats)
Mean, Median, Mode Frequency Table Stdev, Variance Median, Mode Range, IQR Min, Max Skewness, Kurtosis
(All Descriptive Stats)
(Limited Descriptive Stats)
(Limited Descriptive Stats)
Mean, Median, Mode Frequency Table Frequency Table Stdev, Variance Median, Mode Mode Range, IQR Min, Max Skewness, Kurtosis
Summarise Responses
What is your ethnicity?
(please tick one)
White/European Asian West Indian African Other
What is the variable? Level of Measurement? Appropriate numerical summary?
Statistics Frequency tables Charts
Bar Pie Histogram
Representing data Numerically
Nominal Variable
e.g., Ethnicity
Representing data Numerically
Nominal Variable
e.g., Ethnicity
Not really needed. The frequency table shows the mode is White/Euro (most frequent Ethnicity)
Summarise Responses
I am very committed to this organisation.
Strongly Disagree 1 Disagree
2
Neutral
3
Agree
4
Strongly Agree 5
What is the variable? Level of Measurement? Appropriate numerical summary?
Representing data Numerically
Ordinal Variable
e.g., Organisational Commitment
Representing data Numerically
Ordinal Variable
e.g., Organisational Commitment
Not really needed. The frequency table shows the mode.
Representing data Numerically
Out of the total sample size (e.g., 70)
14 out of 70 is 20.0%
Representing data Numerically
Out of the total sample size (e.g., 70)
16 out of 70 is 22.9%
Representing data Numerically
Out of the total sample size (e.g., 70)
68 out of 70 is 97.1%
Representing data Numerically
Out of the total responses (i.e., 68)
14 out of 68 is 20.6%
Representing data Numerically
Out of the total responses (i.e., 68)
16 out of 68 is 23.5%
Representing data Numerically
Adds the Valid Percentages
Same as first valid percent
Representing data Numerically
Adds the Valid Percentages
Adds the 1st two valid percents
Representing data Numerically
Adds the Valid Percentages
Adds the 1st three valid percents
Summarise Responses
What is your gross annual income?
What is the variable? Level of Measurement? Appropriate numerical summary?
Representing data Numerically
Scale Variable
e.g., Gross Annual Income
Select any of these Descriptives
Representing data Numerically
The average income: 7819.12
Representing data Numerically
The middle income: 7800 Half earned less Half earned more
Representing data Numerically
One of the most frequent incomes: 6800
Representing data Numerically
On average, each employees income is approximately 998 different from the mean
Representing data Numerically
The lowest income of all employees: 5900
Representing data Numerically
The highest income of all employees: 10,500
Cross Tabulation Table
use Crosstabs in SPSS
All Descriptive Statistics
Split file then use Frequencies in SPSS
Summarise Responses
I am very committed to this organisation.
Strongly Disagree 1 Disagree 2 Neutral 3 Agree 4 Strongly Agree 5
Gender
(please tick one)
Male
Female
What are the variables?
Level of Measurement? Appropriate numerical summary?
Crosstabs table
(Counts or %s) Observed tallies Expected tallies
Crosstab statistics
Chi-square Correlations
Clustered bar charts
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
Observed count frequency observed in the sample
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
Column % % of participants using column totals
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
Row % % of participants using row totals
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
Total % % of participants using sample size
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
6 Females Strongly Disagreed
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
8 Males Agreed
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
19 participants were Undecided
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
6 of 38 Females Strongly Disagreed 15.8% of females
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
8 of 30 Males Agreed 26.7% of males
Representing data Numerically
Two Categorical Variables
e.g., Gender by Commitment
19 of 68 Participants were Undecided 27.9% of the sample
Summarise Responses
What is your gross annual income?
Gender
(please tick one)
Male
Female
What are the variables?
Level of Measurement? Appropriate numerical summary?
Two methods:
1. Explore option 2. Split file, then Frequencies option We will use the
2nd method.
Method 2
Split the file by Gender
Output for males and females will be in one table
Separate output tables for males and females
File must be sorted!
Categorical Variable
Method 2
Split the file by Gender
The file is sorted All females are listed first, then all males
Method 2
Statistics Frequency tables Charts
Bar Pie Histogram
Now get stats for Annual Income Use Frequencies Output will be split by Gender
Method 2
Representing data Numerically
Scale Variable
e.g., Gross Annual Income
Select any of these Descriptives
Method 2
Representing data Numerically
Split File Option to Compare Groups
Output for males and females is in one table
Method 2
Recall the Split File output options
Output for males and females will be in one table
Separate output tables for males and females
Method 2
Representing data Numerically
Split File Option to Organise Output by Groups
Separate output table for males and females
Method 2
Remember to go back and Analyse all cases, do not create groups
Method 2
Open the data file: 1991 U.S. General Social [Link] Located on the L: Drive in the following folders: CC SPSS1802 Samples English Select 1991 U.S. General Social [Link]
Summarise Responses
Explore the relationship between:
gender age happiness and whether a person finds life exciting
using appropriate numerical and graphical summaries. You decide which variables to compare
SPSS Survival Manual, 4th Edition (2010) by Julie Pallant.
(For SPSS Version 15 or later)
Discovering Statistics Using SPSS, 3rd Edition (2009) by Andy Field.
(For SPSS Version 15 or later)