Question

 

I'm doing a research project for my research methods class in college, and I was confused on this assignment we're doing currently. Right now we've been doing rough drafts of the research paper dealing with different sections without having data. For each section we've really just been talking in a future sense predicting what we think might happen. This section confuses me because We're using Qualtrics for surveys to get data on our research project, and I'm not sure how exactly to say how I'm testing my hypothesis for the descriptive and inferential aspects. My hypothesis is that, if someone is altruistic, then they will be less violent after being made death aware, than those who aren't. The survey consists of making the person death aware, then asking them about a scenario that has a scale on how likely they are to do this action. (There are different variables, but just to save on time that's the gist of the experimental group). How would you suggest I go about saying how I'd give the expected results for the descriptive and inferential parts? Here's the prompt as well

Normally, you would have a Results section in which you describe how you will test your hypothesis using descriptive and inferential statistics. While you will have the opportunity to do this at the end of the semester, we are going to focus on the “big picture” aspects of writing the expected results in your proposal. First, remind the reader of your first hypothesis. Then, outline how you will test that hypothesis statistically (including descriptive statistics, inferential statistics, and graphically). How would you know if your hypothesis is supported (if a one-tailed hypothesis is predicted, you would check your means and the p-value; if only a two-tailed hypothesis, the p-value is all you need to determine statistical significance.

Answer 1

It is advisable to assign some sort of score to the likelihood of the subjects committing a violent act before and after being made death aware. Let's assume this score is determined simply by the number of 'Yes' said to the questions in the survey (assuming the questions are like: Are you likely to hurt someone, etc), before and after being made death aware.

This needs to be done on independent samples using some sort of random sampling technique, using one sample before making the subjects death aware and the other sample after making the subjects death aware. The sampling can be repeated and the mean scores from the before- and after- surveys can be plotted to understand the change. The standard deviation of the scores can also be plotted to understand if there is some sort of increased divergence in the results of the before- and after- surveys.

Now, coming to the inferential part, the hypothesis can be laid out as:

H₀: Being made death aware does not change the likelihood of the subjects committing the violent act (), where subscript b stands for before and a for after)

H_a: Being made death aware does change the likelihood of the subjects committing the violent act ()

This hypothesis can be tested using the 2-sample equality test for population means.

Say, we are testing the validity of the alternative hypothesis at the 5% significance level. First, we calculate the Z-statistic for the given samples, and then the p-value p' for this Z-statistic using the Z-score table. If the p-value p' is less than the significance level 0.05, we reject the null hypothesis, at the 5% significance level, establishing that there is sufficient evidence to suggest that the process of making the subjects death aware does indeed reduce their likelihood of committing the violent acts.