Develop a research project that uses variables you can measure with items in the Statistics Canada General Social Survey: Social Identity dataset. Explain what measures you will use to answer your question, and prepare them for analysis. Your project must be sociological or anthropological in nature, and rely on measures from the dataset. Complete the following. Steps 1-3 focus on conceptualization; steps 4-5, on operationalization. To turn in your write-up as a Word file, your edited dataset, and your SPSS output saved as a PDF file. 1. State your topic. 2. List and define or describe each of the three variables that your research question engages with, in one sentence or less for each. State which one is your outcome variable (OV) and which are your two predictor variables (PV). Remember that you should have an indicator in the dataset that you can use as a measure for each of your variables. See part 4 for level of measurement requirements for the measures. 3. What relationship do you expect to see between each of your predictor variables and your outcome variable? A. State two research questions involving your variables. You should have two separate bivariate research questions. The first should ask about the relationship between your first PV and your OV, and the other about the relationship between your second PV and your OV. See reminder examples on the next page. If you want to practice thinking about multivariate research questions, you can optionally pose one here. B. Draw a diagram showing the relationships you expect to see between your three variables (You can use Word or do it by hand and attach it.) C. In 2-4 sentences, explain why you find the topic and questions interesting or important your measures. Your OV, and one PV, should be interval, ordinal-treated-as- interval, or ratio. The other PV should be nominal/categorical (or a variable you will code as categorical with two or three categories). After completing data cleaning for each measure, list the… (See example assignment for formatting options.) 1) Original name of measure (from codebook/SPSS), and final name if you changed it. 2) Original label for measure (from codebook/SPSS), and final label, if you changed it. 3) Original and final coding schemes. 4) Final level of measurement (might be the same as the original level of measurement). 5) Describe the data cleaning you completed in your SPSS dataset. A reminder of the steps: Check the coding scheme and conduct any necessary recoding, make a new name and descriptive label if necessary, specify missing values and the correct level of measurement. 6) For recoded measures, include the crosstab confirming correct recoding. 5. Evaluate the match between each of your variables and the measures you are using. Are there any limitations? Have you had to adjust your conceptualization because of the available measures? Can you think of what a better measure might look like? Your measure might not exactly match your initial conceptualization of the variable, but it should be a reasonable fit. 5. Evaluate the match between each of your variables and the measures you are using. Are there any limitations? Have you had to adjust your conceptualization because of the available measures? Can you think of what a better measure might look like? Your measure might not exactly match your initial conceptualization of the variable, but it should be a reasonable fit. Research topic is sociological or anthropological. Research questions are clearly stated, and the reasoning for their being interesting is relevant. Variables are appropriate for the research question. Variable definitions are clear. Measures are correctly identified and matched with the correct level of measurement, etc. Appropriate changes are completed successfully in the SPSS dataset and reported in the assignment. The match between concepts and measures is appropriately and clearly evaluated. Any limits to the match are pointed out and evaluated for how severe they are. The fit is reasonable.
