Experimental Design Statistical Analysis of Data Assignment

Assignment Task

Description

You will be given the dataset from an online survey we will conduct as a class. Your task is to, individually, examine the items in the questionnaire, and come up with some hypothetical constructs that you believe some of the items may be measuring, as well as a research hypothesis about those constructs that you could test using a standard multiple regression. It is expected that each student will identify a completely different set of constructs and a different hypothesis, from looking at the data set. Both your IVs and DVs must be made from scales (created from as series of items that are related).

You must appropriately name your scales, so they are clearly measuring the construct of interest. You should compute these scales to measure the constructs you have identified, and then use a standard multiple regression to test your hypothesis. That is, the analysis should be limited to a basic simultaneous regression – not hierarchical, mediated, or moderated. Use simple Cronbach’s alpha calculations to check the reliability of your scales – do NOT perform a factor analysis. You must present all descriptive and inferential statistics that are needed/appropriate. You will write a short Introduction, Method section, Results, and Discussion.

The Introduction and Discussion need not contain any references and should include only a few references at most (3 or 4). Note there are no grades awarded for research, so do not spend a lot of time reading around the topic. Instead, your Introduction should set out a logical basis for your hypotheses based on common sense (i.e., everyday knowledge that anyone could reasonably possess). Being able to lay out an argument from first principles is an important and often neglected writing skill. The hypotheses should follow logically from your argument, and the choice of analysis should reflect this logic.

Whatever figures or tables are appropriate for your chosen analysis. It should describe whatever assumption checks and diagnostics were performed, succinctly, any important considerations arising from those checks, and the reasoning behind the choice of the analysis presented, if appropriate

You should know that the aim of all statistical writing (in a thesis or a journal article) is to tell a story about your data. You can consult other sources for inspiration as to the usual style in which regressions are presented, but you will discover that all real results sections are written completely differently to one another – when there is a different story to tell, you must use different words and style to tell it properly. The commonalities are writing should always be succinct and factual, but sufficient to describe everything important about the data. All important claims must be backed up by a significance test, and a measure of effect size where one is available and appropriate. Where something forms the basis of your interpretation of the analysis, it needs to be described in English as well as using numbers

Remember also that copying any more than a few words from any source (including a results section) is plagiarism, and is an extremely serious form of academic misconduct. So, if you do look at other sources for inspiration use them to see the ‘kinds’ of things people include, not the words they use. You have been warned

1) Participate in the survey made available by the UC, via a Canvas Announcement. Participation is voluntary.

2) Download the dataset here.

3) Examine and clean the dataset where necessary.

4) Create scales for your IVs and DV.

5) Perform appropriate statistical analyses relevant to multiple regression.

This Statistics has been solved by our PHD Experts at My Uni Paper.