Indexing and Inference (HW2)
Date Assigned: Week 5
Due Date: Week 6
In completing this assignment, you are to construct an index measuring one dependent variable and then choose indicators for three independent variables, at least one of which has three or more categories; describe all four variables using univariate statistics; hypothesize expected relationships between your dependent and independent variables; and, as a means of assessing the relationships, report the results of three presentation quality crosstabulations.
For this assignment you may use data from any of the 2016 0r 2017 PPIC surveys, or the ANES 2016 data . Other data sets such as those from PEW or the World Values Survey may be used only with explicit, written approval from the instructor. As in all assignments, examples provided in labs or lectures may not be used.
Keep in mind to employ the appropriate weighting variable for your data by including a WEIGHT command at the beginning of your syntax.
This assignment requires you to:
1) Find a set of indicators that you think may measure the same dependent variable and test whether these indicators work together empirically to measure the same concept. Be sure to identify your concept and describe missing values and recodes.
2) Create a summary index as a measure of your dependent variable, reporting alpha and alpha if deleted scores. Report appropriate univariate measures of central tendency and dispersion for your un-recoded (raw) index.
3) Recode your indexed dependent variable into three categories (low, medium, high) and recalculate measures of central tendency and dispersion.
4) Select three independent variables (at least one of which has at least 3 categories) expected to relate to the indexed dependent variable. Be sure to state what concepts your independent variables are measuring. Then examine the relationships between your three independent variables and your recoded dependent variable via crosstabulation and appropriate summary measure(s) of association.
5) Use Chi Square to assess the statistical significance of all three relationships and report the extent to which further analysis with one-way ANOVA sheds light on the three crosstabulations.
6) Referring to all of the available evidence as well as broader theoretical considerations, explain which of the independent variables offers the best explanation for variation in the dependent variable and summarize your findings.
7) So that your work may be replicated, please include your syntax as an attachment to your written assignment.