Bivariate analysis in SPSS
SPSS LABORATORY 6

1. Calendar
2. Bivariate analysis: quick theory refresher
3. Bivariate analysis in SPSS
4. Exercise
INDEX

No content

Photo: Startup Weekend Hackathon. Nov.2014
SURVEY
Do not forget to take the survey by Friday, 18 at midnight.
Canvas > Quiz > CMNS201 Survey 2022

Photo: Startup Weekend Hackathon. Nov.2014
BIVARIATE ANALYSIS: DEFINITION
Bivariate analyses are conducted to determine whether a statistical association exists between two variables. For
example, bivariate analyses could be used to answer the question of whether there is an association between
income and quality of life.
Source: Sandilands D.. (2014) Bivariate Analysis. In: Michalos A.C. (eds) Encyclopedia of Quality of Life and Well-Being Research. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-0753-5_222

Photo: Startup Weekend Hackathon. Nov.2014
BIVARIATE ANALYSIS: DEFINITION
More speciﬁcally, bivariate analysis explores how the dependent variable depends by the independent variable.
There are 2 types of relationship between the dependent and independent variable:
■ A direct relationship (also called positive correlation) – that means if the independent variable increases,
then the dependent variable would also increase and vice versa.
■ A inverse relationship (negative correlation) – when the independent variable increases and the dependent
variable decrease and vice versa.

Photo: Startup Weekend Hackathon. Nov.2014
BIVARIATE ANALYSIS: EXAMPLE
Average daily temperature Ice cream sales
Example of a direct relation. If average daily temperature increases, ice cream sales also increase

Photo: Startup Weekend Hackathon. Nov.2014
BIVARIATE ANALYSIS: EXAMPLE
Average daily temperature Ramen sales
Example of an inverse relation. If average daily temperature increases, ramen sales decrease

Photo: Startup Weekend Hackathon. Nov.2014
HOW TO CONDUCT BIVARIATE ANALYSIS IN SPSS
■ Go to Canvas
○ Assignments > SPSS Lab 6 - Bivariate analysis
○ Download the Week-6.sav ﬁle
■ Open the ﬁle on SPSS
○ Launch SPP
■ File > Open > Data…
■ Find and open the Week-6.sav ﬁle

Photo: Startup Weekend Hackathon. Nov.2014
LET’S BEGIN BY FORMULATING A HYPOTHESIS
SFU Students with Canadian citizenship will tend to have higher GPA because they grew up within the Canadian
educational system.
On the contrary, SFU students on a study permit will have lower GPA because in addition to entering a new
educational system, they also experience disconnection from their families and communities.
Null hypothesis: there is no relation between Immigration Status and GPA

Photo: Startup Weekend Hackathon. Nov.2014
LET’S BEGIN BY FORMULATING A HYPOTHESIS
SFU Students with Canadian citizenship will tend to have higher GPA because they grew up within the Canadian
teaching system.
On the contrary, SFU students on a study permit will have lower GPA because in addition to entering a new
educational system, they also experience disconnection from their families and communities.
Immigration status:
Independent variable
GPA:
dependent variable
Null hypothesis: there is no relation between Immigration Status and GPA

Photo: Startup Weekend Hackathon. Nov.2014
BIVARIATE ANALYSIS IN SPSS

Photo: Startup Weekend Hackathon. Nov.2014
SETTING UP YOUR CROSSTAB
Independent variable in the Rows
Dependent variable in the Columns

Photo: Startup Weekend Hackathon. Nov.2014
SETTING UP YOUR CROSSTAB
1. Put Residence_status in the Rows
2. Put GPA in the Columns
3. Click Cells

Select Observed and Expected
counts
Photo: Startup Weekend Hackathon. Nov.2014
SETTING UP YOUR CROSSTAB
Select Row
Click Continue

Photo: Startup Weekend Hackathon. Nov.2014
SETTING UP YOUR CROSSTAB
4. Click Ok

Photo: Startup Weekend Hackathon. Nov.2014
HOW TO READ A CROSSTABULATION
Independent
variable
Dependent variable

Photo: Startup Weekend Hackathon. Nov.2014
HOW TO READ A CROSSTABULATION
Count: is the observed
frequency in a particular
cell of the crosstabs table.
For example, In this case,
18 students with Canadian
Citizenship have a GPA of
2.99 or lower.
Expected Count: The expected
count is the predicted frequency
for a cell under the assumption
that the null hypothesis is true. In
our case, the null hypothesis is
that there is no association
between Immigration Status and
GPA.
If the Null hypothesis was true,
then we expected 17 students
with a GPA of 2.99 or lower.

Photo: Startup Weekend Hackathon. Nov.2014
HOW TO READ A CROSSTABULATION
*: “signiﬁcant” is a very broad qualiﬁer. In general, if you see a diﬀerence equal or greater of
10%, you can assume there is a relation between the two variables (which can conﬁrm or not
your hypothesis). If the diﬀerence is small (around 5%), you can assume there is no relation
between the two variables.
To understand if our initial
hypothesis is conﬁrmed or not,
we need to compare % by
column.
To conﬁrm our hypothesis, the %
of Canadian students with a GPA
greater than 3 should have been
signiﬁcantly* above the % of
study permit students with a GPA
above 3.
This results do not support the
hypothesis. Actually show the
opposite: there is a relation
between Residence_Status and
GPA but a relatively larger share
of Study permit students (4.4%)
compared to Canadian Citizen
appear to have a GPA of 3 or
higher.

Photo: Startup Weekend Hackathon. Nov.2014
CROSSTABULATION AND VARIABLE RECODING
Hypothesis: People who sleep less tend to achieve a higher GPA.
Null hypothesis: there is no relation between Sleep_hrs and GPA

Photo: Startup Weekend Hackathon. Nov.2014
CROSSTABULATION AND VARIABLE RECODING

Photo: Startup Weekend Hackathon. Nov.2014
CROSSTABULATION AND VARIABLE RECODING
8 is usually indicated as the ideal number of hours of sleep per night. Based on this information, let’s create 2 segments.
In formula:
■ If Sleep_hrs <= 7.5, then Sleep_new = “Restless”
■ If Sleep_hrs => 8, then Sleep_new = “Sleepy”

Photo: Startup Weekend Hackathon. Nov.2014
CROSSTABULATION AND VARIABLE RECODING
If you do not remember how to recode a variable, check Week 3 slides.

Photo: Startup Weekend Hackathon. Nov.2014
1: select the
variable to
recode
2: type the new
variable name and
add descriptive label
3: click change
4: click Old and
New Values
5: set your range
6: tick the checkbox
and deﬁne width 15
7: deﬁne the new
variable value for the
range
8: click add and
repeat steps 5-7-8 for
the second range.
10: click OK
9: once deﬁned all
ranges, click Continue

Photo: Startup Weekend Hackathon. Nov.2014
CROSSTABULATION AND VARIABLE RECODING

Photo: Startup Weekend Hackathon. Nov.2014
CROSSTABULATION AND VARIABLE RECODING
The ﬁndings do not support the hypothesis. The % of students with a GPA of 3 and above is nearly identical for both Sleepy and
Restless students.
The diﬀerence is
just 1.5%!

Photo: Startup Weekend Hackathon. Nov.2014
EXERCISE
Run a Crosstab for the variables Gender (independent Variable) and GPA (dependent variable).
Is there a relation between the two variables?
Upload a screenshot of the crosstab on Canvas alongside a one sentence comment about the relation (or lack
thereof) between the two variables.

THANK YOU
Alberto Lusoli
[email protected]
Oﬃce hour: Thursday, 12.30pm - 1.20pm (please book an appointment in advance via email).