Lab 1

POL242 LAB MANUAL: EXERCISE 1

 

PURPOSE

  • Learn how to use variables and their indicators.
  • Become familiar with searching out indicators for variables in codebooks.

 

MAIN POINTS

  • Social Research entails systematic observation and the understanding of patterns. Systematic observations are made using surveys, experiments, available data and field research. In this exercise we will use survey data. Patterns are understood by identifying and explaining variation in our observations. An essential aspect of social research is making certain that our observations exhibit some degree of variation. If there is little or no variation in our observations, they are regarded as constants. Constants are of limited value for most social research. So we try to look for variation.
  • Variables are abstract concepts used to describe sets of attributes or characteristics. Some simple examples are gender, age and race, but social researchers are also interested in more complex variables such as political identifications or support for immigration or security policies.
  • Indicators are used to measure variables. For example, to measure political identification we could use the indicator, “Generally speaking do you think of yourself as a Liberal, Conservative, New Democrat or Green ?”  To measure support for immigration we might use the indicator, “From what you can tell, do you think Canada accepts too many, too few or about the right number of immigrants per year?”

Stage I

INSTRUCTIONS

  1. Find a Forum Research Poll at Dataverse for this first exercise. These data were made available for student use by Forum Research.
    They are available at: dataverse.scholarsportal.info/dvn or (when off campus) through:

    • https://login.library.utoronto.ca/cgi-bin/go_log.pl?url=http://dataverse.scholarsportal.info/dvn/dv/MDL
  2. Search for the Forum Polls by entering the word Forum in the search box.
  3. Select either the Federal, Provincial or Municipal Polls.
  4. Click on the Data & Analysis link.
  5. Download the codebook for any of the multiple data sets available.
  6. Find a question that is an indicator of a variable that is of interest to you.
  7. Download as an SPSS file the data set corresponding to the codebook containing the variable in which you are interested.
  8. Before opening the data set, select Save As and save the file on your desktop of another suitable spot.
  9. Right click on data set icon, select Properties and lock the dataset by clicking the box labeled “Read Only.” On a Mac right click on the icon and select Get Info and lock the dataset by clicking the box labeled locked
  10. Open the dataset using SPSS.
  11. Under the Analyze menu, select Descriptive Statistics and then Frequencies.
  12. Locate the item you have chosen for analysis and move it into the Variables area.
  13. Click Paste and review the syntax
  14. Select the syntax.
  15. Go either to the Run menu and choose the green triangle or click on the green triangle in the row of icons.
  16. Review the results and ask yourself: Is there variation on this question?
  17. If there is variation try to think of an explanation for the variation you observe by looking at the other available questions in the survey.
  18. Formulate a hypothesis in the form of X –> Y.
  19. Repeat steps 1-6 to examine variation on the hypothesized explanatory (X) variable.

EXAMPLE (using Forum data)

  • Questionnaire
    • Forum Provincial Issues 2013-12
  • Y Variable
    • Attitudes regarding wind power
  • Indicator for Y
    Q8: Do you approve or disapprove of windfarms, or groups of windturbines in rural areas?
  • Possible Explanation (X)
    Education
  • Indicator for X
    D4: What is the highest level of education you completed?
  • SPSS Syntax

    Fre var q8, d4.
  • QUESTION FOR REFLECTION
    Is there variation on both the dependent and independent variables that form your hypothesis?

 

Stage II

MAIN POINTS

  • A number of different data sets are available for analysis.
  • More complex data sets allow greater scope for investigation.
  • The CES 2011 and the ANES 2012 offer a wider range of variables to explore. Links to both data sets and their respective codebooks are available via the course website.
  • Other studies are also available though you will be required to use particular data sets for your homework assignments. Please consult the relevant assignments for further details.

INSTRUCTIONS

  1. Repeat the exercise in Stage I using either the CES 2011 or the ANES 2012.
  2. Find the Questionnaire for the dataset you have chosen by following the links under the first week’s reading from the POL242 website.
  3. Think of a simple variable that is related to the topic of your chosen dataset.
  4. Find several indicators of the variable you have selected in the questionnaire
  5. To assist in your search, you may wish to use the Find function in the menu bar under Edit. When using the find function, you may have to enter synonyms of your chosen issue to find appropriate variables.

EXAMPLE (using CES 2011)

  • Questionnaire
    • CES 2011
  • Y Variable
    • Attitudes regarding reducing inequality
  • Indicator for Y1
    • PES11_41: How much should be done to reduce gap between the rich and the poor in Canada?
  • Additional possible indicators for Y (attitudes regarding reducing inequality Y2-4):
    mbs11_k2 – the government SHOULD ACT/SHOULD NOT ACT to reduce differences in income and wealth;
    mbs11_b3 – the govt should see to it that everyone has a decent standard of living/leave people to get ahead on their own.
    pes11_52b – an NDP government would really hurt the Canadian economy.
  • Possible Explanation (X1)
    Party Identification
  • Indicator for X1
    CPS11_71: In federal politics, do you usually think of yourself as a: Liberal, Conservative, N.D.P, Bloc Québécois, Green Party, or none of these?

SPSS Syntax (note: commands between asterisks are commentary and not executed by SPSS)

*Preparing a Y indicator-Cdn Attitudes re Inequality*
missing values pes11_41 (8,9).
recode PES11_41 (1=1) (2=.75) (3=.5) (4= .25) (5=0) into undogap.
value labels undogap 0 'muchless' .25 'someless' .5 'asnow' .75 'somemore' 1 'muchmore'.

*frequency analysis for Y*.
fre var undogap.

*Preparing the X indicator- party identification*.
recode cps11_71 (2=1) (1=2) (4=3) (3=4) (5=5) (6 =6) into PID.
value labels PID 1 'Cons' 2 'Lib' 3 'BQ' 4 'NDP' 5 'Green' 6 'None'.

*frequency analysis for Y1*.
fre var PID.

Formatted Tables Produced from Output

How much should be done to reduce gap between the rich
and the poor in Canada?

Value Labels Frequency Percent Cumulative %
Much Less 63 1.5 2.0
Somewhat Less 81 1.9 4.5
Same as Now 638 14.8 24.2
Somewhat More 1252 29.1 63.0
Much More 1193 27.7 100.0
Total 3227 74.9

Source: CES 2011

 

Respondents’ Party Identification

Value Labels Frequency Percent Cumulative %
Cons 1159 26.9 29.3
Lib 1040 24.1 55.5
BQ 357 8.3 64.5
NDP 452 10.5 75.9
Green 102 2.4 78.5
None 852 19.8 100.0
Total 3962 100

Source: CES 2011

Conclusion
There is variation of the indicators of both X and Y.

FOR FURTHER REFLECTION (The following syntax contains some advanced techniques)

Syntax for Additional Indicators of Y (Y2-4).

*These commands reverse the order and relabel the Y2 indicator which has values running from 0-10*.
missing values mbs11_k2 (-99).
compute govact = (((mbs11_k2 * -1) +10)/10).
value labels govact 0 'not act' 1 'gov act'.

*The next four commands recode and relabel the Y3 and Y4 variables*.
recode mbs11_b3 (1=1) (2=0) into goveqch.
value labels goveqch 1 'decent living' 0 'leave alone'.

recode pes11_52b (1=0) (3= .33) (5= .66) (7= 1)into NDPnohurt.
value labels NDPnohurt 0 'strdisagree' .33 'disagree' .66 'agree' 1 'stragree'.

*frequency analysis of Y2-4*.
fre var govact goveqch NDPnohurt.

Some additional explanatory variables (X2-5) of possible interest.

cps11_78  (age)
rgender11  (gender)
cps_intlang11 (language)
cps11_79  (education)

Syntax for additional explanatory variables X2-5* of possible interest.

*Create Age indicator*.
missing values cps11_78 (9998, 9999).
compute age = (2011- cps11_78 ).

*Create gender indicator*.
recode rgender11 (1=0) (5=1) into female.

*Create Language indicator*.
recode cps_intlang11 (1=0) (5=1) into french.

*Some more complex measures X6-7*.
*Create education measure*.
recode cps11_79 ( 1=1) (else = copy) into educ.
value labels educ 1 ' no sch' 2 ' some elem' 3 'compl elem' 4 'some sec'
5 'compl sec' 6 'some tech cc' 7 'compl tech cc' 8 'some univ'
9 ' bach' 10 ' mast' 11 'prof doc' 98 'dk' 99 'ref'.
missing values educ (98,99).
recode educ (1 thru 4 = 1) (5=2) (6,7,8=3) (9=4) (10,11=5) into educ5.
value labels educ5 1 '<sec' 2 'sec' 3 'some post sec' 4 'post sec' 5 'grad sch'.

*Create Income measure*.
missing values cps11_92 (998, 999).
missing values cps11_93 (98,99).
numeric income.
if (cps11_93 =1) or (cps11_92 lt 30) income = 1.
if (cps11_93 =2) or ((cps11_92 ge 30) and (cps11_92 lt 60)) income = 2.
if (cps11_93 =3) or ((cps11_92 ge 60) and (cps11_92 lt 90)) income = 3.
if (cps11_93 =4) or ((cps11_92 ge 90) and (cps11_92 lt 110)) income = 4.
if (cps11_93 =5) or (cps11_92 ge 110) income = 5.
value labels income 1 '<$30k' 2 '$30k-$59k' 3 '$60k-$89k' 4 '$90k-$109k' 5 '$110k+'.

fre var age, female, french, educ, income.