Create a table(s) which describes the distributions of your independent, dependent, and mediator/confounder variables.

Sociological report

Form a sociological hypothesis/ research expectation about inequality in health, pensions, health behaviours and well-being and their link to stratification variables such as class, ethnicity, and gender.

1. You should aim to build on your Report plan using data from Lab sessions in weeks 23 and 25, the Understanding Society survey, however, it can be possible that your thinking has evolved, and you instead choose other data – describe in a sentence or two why you needed to use other data.

2. Link the hypothesis(es)/research expectations and to existing literature (from class and/or additional sociological scholarship)

3. Write a brief Literature review that sums-up what previous studies have done, max 300-400 words

4. Positapotential mediator/moderator in this relationship.You are encouraged to posit a moderator as this is more challenging, and extra points will be given for attempting this. If not, remember that we learned to use mediators in Lab Practical Week 23.

5. Describe the data that you will be using and the variables you are going to focus on, 150-max 200 words. If needed, post a link to the where the data can be found there or from other websites from which it can be downloaded). We usually include such sections in papers because they mean that our analysis can be replicated.

B. Test your hypothesis with data
1. Create a table(s) which describes the distributions of your independent, dependent, and mediator/confounder variables. It is important to demonstrate:

i. appropriate variable choice and coding (*return to previous labs for ways to recode. Remember dummy variables?).

ii. Appropriate format of descriptive statistics and/or frequencies (depending on the nature of your variable).

iii. Correct interpretation of table. Look for values in the min, max, and mean and what they say about your variables.

2. Create a table of linear regression results
i. At least two models: (1) simple or multiple regression (extra marks for multiple regression!), (2): 2 + one mediator or moderator main effect.

Remember to make explicit to the reader what your hypothesis was and reinstate, if needed, what effect you will expect from a moderator or mediator.

If multiple regression is chosen, look at Lab practical Week 25.

If a mediator is chosen, look at Lab practical Week 23.

If a moderator is chosen (extra marks), look at Lab practical Week 25 (pages 7-8). Here you will also have to provide results for a third model that includes for the interaction effects.

ii. Correct interpretation of each model, and changes in coefficients between models (remember to look at your last two lab sessions for correct step-by- step guides to perform and interpret results).

C. Drawing conclusions from your work:

1. Is your hypothesis supported?

2. Can you provide an interpretation of why or why not? You should aim to have about 10-15 references in total.

Create a table(s) which describes the distributions of your independent, dependent, and mediator/confounder variables.
Scroll to top