STATISTICS
SPSS
Choose for result section
How to get output/ SPSS functions
What to report
Where to look in the output
,Examples hypothesis
Linear Regression (predict)
1 IV (scale) + 1 DV (scale)
1. A person’s income can predict reported happiness.
2. A person’s score on the BAS_drive scale predicts how much alcohol they drink.
3. A person’s BIS score (behavioral inhibition) predicts how worried they are about failing this course.
Correlation (relationship)
1 IV (scale) + 1 DV (scale)
1. Is there a relationship between years of education and happiness?
2. Students that drink less alcohol are more conscientious.
3. Students from a larger city drink more alcohol.
Partial correlation
Now that I think about it, perhaps there is another variable that might influence the relationship between amount
of education and happiness. Test whether being married or single affects the relationship between years of
education and happiness.
One-samples t-test
1. testing whether a score is higher than the mid-point of the scale
Independent t-test
1 IV (nominal, 2 levels/groups) + 1 DV (scale)
1. Friends enjoy the game more than strangers.
2. Do men drink more units of alcohol per week than women?
3. Do women spend more hours per course than men?
Dependent (paired) samples t-test
1 IV (1 level/groups, 2 times) + 1 DV (scale)
,1. For this question, we will look at whether some words are easier to explain than others. Test the following
hypothesis: It takes less time to explain the word ‘string’ than the word ‘trampoline’.
Include a figure to illustrate the pattern of means.
2. People’s correctness judgment of real English words is better than their correctness judgment of non words.
One-way ANOVA
1 IV (nominal, but more than 2 levels) + 1 DV (scale)
1. The type of relationship determines how much closeness people report. Specifically, people report more
closeness to someone they know than to a strangers and people report more closeness to a friend than to an
acquaintance
2.. Test whether education level influences happiness. People with lower education should be less happy than
those with middle education, who in turn should be less happy than those with higher education.
3. There is a difference in how much students drink. Specifically, students who do a business related bachelor drink
more than students doing a media or language related bachelor. Moreover, students who do a media related
bachelor drink more than students who do a language related bachelor.
Two-way ANOVA / Factorial ANOVA
2 IVs (both are nominal) +1 DV (scale)
1. The role (explainer, guesser) influences how difficult people think the game was. Those who explain think it is
more difficult than those who guess.
2. Ads with image animation score higher on attitude than ads without image animation (main effect). Ads with a
tagline score higher on attitude than ads without a tagline (main effect). For ads with animation the score on
attitude more strongly depends on the presence of a tagline (interaction)
3. More expensive gifts are appreciated more than cheaper gifts. Specifically, an invitation to a cheap restaurant is
appreciated less than an invitation to a middle-class restaurant, which is in turn appreciated less than an invitation
to a gourmet restaurant.
Women appreciate an invitation to a restaurant more than men. For women, the appreciation of the gift more
strongly depends on the price of the gift than for men.
Chi-square
1 IV (nominal) + 1 DV (nominal)
1. Friends are more likely to use shared knowledge than acquaintances.
2. Boys are more likely than girls to have their own mobile phone.
3. Children from divorced parents are more likely to have a mobile phone than children whose parents are not
divorced
, 4. Males are more likely to place watching TV on first place as their favorite way to spend leisure time than
females.
TEST EXPLAINING
Correlation
1 IV (scale) + 1 DV (scale)
A correlation was used to examine the relationship between ordinal, interval or ratio variables
(= continuous variables, but NEVER the relationship between nominal variables, because the values
always have to have a meaningful order for correlation).
Specifically, it measures to what extent two variables are related. It does not say anything about
causality. It measures the pattern of responses across variables.
The DV needs to be continuous. In general, both variables need to be continuous for a correlation.
There is a special case in which the IVs categorical, leading to a point-biserial correlation. The IV needs
to be dichotomous, with only two levels (e.g. male/female) for this to apply. In this special case, the IV
does not need to be normally distributed, of course, because a dichotomous variable cannot be
normally distributed.
Partial Correlation
Measure the relationship between two variables, controlling for the effect that a third variable has on
them both.
• Helps to deal with the third-variable problem, but cannot solve it (there may be more)
• E.g., exam anxiety, exam performance & revision time
Linear Regression
Linear regression (depends on the hypothesis if it just wants to check the relationship or how IV can
predict DV)
A linear regression was used to predict one variable from another, that is, when we want to see if one
variable predicts the behaviour of another one. More precisely, if X and Y are two related variables, then
linear regression analysis helps us to predict the value of Y for a given value of X or vice versa.
Unlike a correlation, it says something about the direction of the relationship. It can only be used when
there is one independent variable, and the dependent variable must be continuous. It aims to determine
how well a model generalizes to the population > allows to confirm or reject the hypothesis.
Simple regression
SPSS
Choose for result section
How to get output/ SPSS functions
What to report
Where to look in the output
,Examples hypothesis
Linear Regression (predict)
1 IV (scale) + 1 DV (scale)
1. A person’s income can predict reported happiness.
2. A person’s score on the BAS_drive scale predicts how much alcohol they drink.
3. A person’s BIS score (behavioral inhibition) predicts how worried they are about failing this course.
Correlation (relationship)
1 IV (scale) + 1 DV (scale)
1. Is there a relationship between years of education and happiness?
2. Students that drink less alcohol are more conscientious.
3. Students from a larger city drink more alcohol.
Partial correlation
Now that I think about it, perhaps there is another variable that might influence the relationship between amount
of education and happiness. Test whether being married or single affects the relationship between years of
education and happiness.
One-samples t-test
1. testing whether a score is higher than the mid-point of the scale
Independent t-test
1 IV (nominal, 2 levels/groups) + 1 DV (scale)
1. Friends enjoy the game more than strangers.
2. Do men drink more units of alcohol per week than women?
3. Do women spend more hours per course than men?
Dependent (paired) samples t-test
1 IV (1 level/groups, 2 times) + 1 DV (scale)
,1. For this question, we will look at whether some words are easier to explain than others. Test the following
hypothesis: It takes less time to explain the word ‘string’ than the word ‘trampoline’.
Include a figure to illustrate the pattern of means.
2. People’s correctness judgment of real English words is better than their correctness judgment of non words.
One-way ANOVA
1 IV (nominal, but more than 2 levels) + 1 DV (scale)
1. The type of relationship determines how much closeness people report. Specifically, people report more
closeness to someone they know than to a strangers and people report more closeness to a friend than to an
acquaintance
2.. Test whether education level influences happiness. People with lower education should be less happy than
those with middle education, who in turn should be less happy than those with higher education.
3. There is a difference in how much students drink. Specifically, students who do a business related bachelor drink
more than students doing a media or language related bachelor. Moreover, students who do a media related
bachelor drink more than students who do a language related bachelor.
Two-way ANOVA / Factorial ANOVA
2 IVs (both are nominal) +1 DV (scale)
1. The role (explainer, guesser) influences how difficult people think the game was. Those who explain think it is
more difficult than those who guess.
2. Ads with image animation score higher on attitude than ads without image animation (main effect). Ads with a
tagline score higher on attitude than ads without a tagline (main effect). For ads with animation the score on
attitude more strongly depends on the presence of a tagline (interaction)
3. More expensive gifts are appreciated more than cheaper gifts. Specifically, an invitation to a cheap restaurant is
appreciated less than an invitation to a middle-class restaurant, which is in turn appreciated less than an invitation
to a gourmet restaurant.
Women appreciate an invitation to a restaurant more than men. For women, the appreciation of the gift more
strongly depends on the price of the gift than for men.
Chi-square
1 IV (nominal) + 1 DV (nominal)
1. Friends are more likely to use shared knowledge than acquaintances.
2. Boys are more likely than girls to have their own mobile phone.
3. Children from divorced parents are more likely to have a mobile phone than children whose parents are not
divorced
, 4. Males are more likely to place watching TV on first place as their favorite way to spend leisure time than
females.
TEST EXPLAINING
Correlation
1 IV (scale) + 1 DV (scale)
A correlation was used to examine the relationship between ordinal, interval or ratio variables
(= continuous variables, but NEVER the relationship between nominal variables, because the values
always have to have a meaningful order for correlation).
Specifically, it measures to what extent two variables are related. It does not say anything about
causality. It measures the pattern of responses across variables.
The DV needs to be continuous. In general, both variables need to be continuous for a correlation.
There is a special case in which the IVs categorical, leading to a point-biserial correlation. The IV needs
to be dichotomous, with only two levels (e.g. male/female) for this to apply. In this special case, the IV
does not need to be normally distributed, of course, because a dichotomous variable cannot be
normally distributed.
Partial Correlation
Measure the relationship between two variables, controlling for the effect that a third variable has on
them both.
• Helps to deal with the third-variable problem, but cannot solve it (there may be more)
• E.g., exam anxiety, exam performance & revision time
Linear Regression
Linear regression (depends on the hypothesis if it just wants to check the relationship or how IV can
predict DV)
A linear regression was used to predict one variable from another, that is, when we want to see if one
variable predicts the behaviour of another one. More precisely, if X and Y are two related variables, then
linear regression analysis helps us to predict the value of Y for a given value of X or vice versa.
Unlike a correlation, it says something about the direction of the relationship. It can only be used when
there is one independent variable, and the dependent variable must be continuous. It aims to determine
how well a model generalizes to the population > allows to confirm or reject the hypothesis.
Simple regression
