INSTRUCTIONS FOR ASSIGNMENT
FOR SPSS:
Go to: Analyze>Compare Means>One-Way ANOVA
To include the Turkey test, click on Post-Hoc from the dialog box that appears when you go to ANOVA, after you enter the variables (one variable and one factor), click on “Turkey,” click “continue,” then OK
For PSPP
Read the PSPP Chapter 6 for step by step help (Mac Only).
For windows PSPP 1.53
Go to: Analyze>Compare Means>One-Way ANOVA
To include the Turkey test, click on Post-Hoc from the dialog box that appears when you go to ANOVA, after you enter the variables (one variable and one factor), click on “Turkey,” click “continue,” then OK
Test whether there is an association between a person’s educational attainment and how much television they watch. Use the GSS2008 data set to perform an ANOVA on respondents’ highest educational degree (DEGREE) and the hours per day they watch television (TVHOURS)
FOR PSPP ON MAC ONLY: Now perform the analysis with PSPP [USE THE SYNTAX BELOW TO DO THE ENTIRE ANALYSIS] and find:
WE WILL USE THIS TO DO THE FOLLOWING QUESTIONS: One-way ANOVA
File
New
Syntax
The syntax editor will open, then type [do not type in BRACKETS]
Oneway [enter]
/Variables = var1 BY var2 [enter]
/Statistics = descriptives [enter]
/Posthoc = tukey [go to tool ribbon “File Edit Run Windows Help” and click on RUN and then ALL. Check you output and you should have a nice TABLE. Var1 and var2 are the variables being used for the tables. Just type in the variable name as written in the assignment.
Do not type Var1 and Var2 as they do not stand for anything-and you will get an error message.
Hint on the above Tvhours (dependent variable) BY degree (factor)
NOTE CONCERNING SPSS/PSPP DISCUSSION AND ASSIGNMENT EXERCISES: For anyone wondering, I have borrowed most of the exercises from various textbooks that I have used to teach this course and similar courses over the years with some modification when and if necessary AND these questions often challenge the perceptions of students. Part of learning to analyze and interpret data is to set aside ones own biases and perceptions when examining the results.
Read the PSPP ANOVA for step by step help.
1) Astrologers assert that our birth dates influence our success (or lack thereof) in life. Test this assumption with the GSS2018 data by analyzing the relationship between respondent’s astrological sign (ZODIAK) with respondent’s socioeconomic index (SEI10). The SEI10 is an indicator of economic and socio-economic attainment: The higher the SEI score, the more successful the respondent.
Make a prediction:
Can a person’s astrological sign predict their socioeconomic success? Yes No
If so, what zodiac sign is the most successful ____________
Now perform the analysis using the GSS2018 data
Mean SEI10 of Pisces ______________
Mean SEI10 of Taurus ______________
Mean SEI10 of your astrological sign ______________
ANOVA significance level ______________
Is the relationship statistically significant? Yes No
On the basis of these data, would you say that astrologers are correct in their assertion of the power of the stars? Remember to explain this relationship based on what the test measures. Why?
FOR THE REMAINING EXERCISES SEE ABOVE FOR THE SYNTAX SEQUENCE (IN THE WALK THROUGH) BUT MAKE SURE YOU HAVE THE CORRECT VARIABLE NAMES AND IN THE PROPER ORDER
2) Test whether there is an association between a person’s educational attainment and how much television they watch. Use the GSS2018 data set to perform an ANOVA on respondents’ highest educational degree (DEGREE) and the hours per day they watch television (TVHOURS)–Compare with 2008-are the results the same? If not, how where they different?
Make a prediction:
Predict the number of hours each group watches television in a given day:
Less than High School Some College Graduate
High School Grad College Grad Degree
__________ __________ __________ __________ __________
Now perform the analysis with PSPP and find:
Mean hours for people with less than a high school degree ___________________
Mean hours for people with a high school degree ___________________
Mean hours for people with a junior college degree ___________________
Mean hours for people with a bachelor’s degree ___________________
Mean hours for people with a graduate degree ___________________
ANOVA significance level ___________________
Is the relationship statistically significant Yes No
According to the Tukey Test, which categories have means that are significantly different from the means of those with a high school degree?
3) Test whether there is an association between a person’s race and the prestige of their occupation. Use the GSS2018 data set to perform an ANOVA on RACE and PRESTG10. You are to do a tukey test at the end of this one as well so it is easier to just run the syntax like you did for exercise 2 and just change the relevant variables.
Mean prestige score for Whites ___________________
Mean prestige score for African Americans ___________________
Mean prestige score for other races ___________________
ANOVA significance level ___________________
Is the relationship statistically significant Yes No
According to the Tukey Test, which categories have means that are significantly different from the means of Whites?
On the basis of these data, would you say that race is associated with occupational prestige? Remember to explain this relationship based on what the test measures. What could explain this relationship?
4) Test whether there is an association between a person’s degree and the years at current job. Use the GSS2018 data sets to perform an ANOVA on DEGREE and YEARSJOB. You are to do a tukey test at the end of this one as well so it is easier to just run the syntax like you did for exercise 2 and just change the relevant variables..
Make a prediction:
Predict the mean number of years at current job for each group:
Less than High School Some College Graduate
High School Grad College Grad Degree
__________ __________ __________ __________ __________
Now perform the analysis with PSPP and find:
Mean years for people with less than a high school degree ___________________
Mean years for people with a high school degree ___________________
Mean years for people with a junior college degree ___________________
Mean years for people with a bachelor’s degree ___________________
Mean years for people with a graduate degree ___________________
ANOVA significance level ___________________
Is the relationship statistically significant Yes No
According to the Tukey Test, which categories have means that are significantly different from the means of those with a high school degree?
5) Test whether there is an association between a person’s race and the HOURS WORKED LAST WEEK (HRS1). Use the GSS2018 data set to perform an ANOVA on RACE and HRS1. You are to do a tukey test at the end of this one as well so it is easier to just run the syntax like you did for exercise 2 and just change the relevant variables..
Mean HOURS WORKED for Whites ___________________
Mean HOURS WORKED for African Americans ___________________
Mean HOURS WORKED for other races ___________________
ANOVA significance level ___________________
Is the relationship statistically significant Yes No
According to the Tukey Test, which categories have means that are significantly different from the means of Whites?
On the basis of these data, would you say that race is associated with hours worked last week? Remember to explain this relationship based on what the test measures. What could explain this relationship?
6) Test whether there is an association between a person’s race and the YEARS AT CURRENT JOB. Use the GSS2018 data set to perform an ANOVA on RACE and YEARSJOB. You are to do a tukey test at the end of this one as well so it is easier to just run the syntax like you did for exercise 2 and just change the relevant variables..
Mean years for Whites ___________________
Mean years for African Americans ___________________
Mean years for other races ___________________
ANOVA significance level ___________________
Is the relationship statistically significant Yes No
According to the Tukey Test, which categories have means that are significantly different from the means of Whites?
On the basis of these data, would you say that race is associated with years at a current job? Remember to explain this relationship based on what the test measures. What could explain this relationship?