POLS0010

Data Analysis

Term 1, Part 2

ESSAY QUESTIONS 2022–2023 (Term 1 – Part 2)

Guidelines for Completing and Submitting POLS0010 Essay

▪   Read the below guidelines to avoid losing unnecessary marks.

▪   The assessment is due on  11.01.22 at 14.00 hours. Please follow all designated Department  of  Political   Science  submission  guidelines.  THESE  MAY  BE DIFFERENT TO THOSE OF YOUR HOME DEPARTMENT. The submission guidelines are available on the Moodle page for this module. You must submit one copy of your essay via Turnitin. The word limit is 1,500 words, excluding tables and graphs, references, and your R script appendix (see below).

▪   This is an assessed piece of coursework for the POLS0010 module; collaboration and/or discussion with anyone is strictly prohibited. The rules for plagiarism apply and any cases of suspected plagiarism of published work or the work of classmates will be taken seriously.

▪   The data come from the National Surveys of Sexual Attitudes and Lifestyle, 2010-

12 – a face to face survey of 15,162  adults aged  16-74 years. The data  are clustered and stratified probability samples of postal addresses in the UK.

▪   These are real research data from the UK Data Service. The respondents agreed for their data to be used for research and learning purposes. Before you can access this data, you need to agree to some important terms of use. The files you require are contained in the Datasets folder on Moodle.

▪   Once you have agreed to the conditions of use, you may open the dataset and work on the essay questions anytime up until the submission date. There is no limit on the number of times you may open the data file. Be sure to save your data file and R script file regularly.

▪   The essay questions comprise two sections; you must complete each part of each section.

▪   Where appropriate, answers should be written in complete sentences; no bulleting or outlining. Be sure to answer all parts of the questions posed and interpret output statistically and substantively.

▪   You should include tabular and graphical output alongside your written answers in Part A and without any tabular and graphical output in Part B (see below).

▪   You should include a copy ofyour R script as an appendix to your essay. FAILURE TO INCLUDE THE R SCRIPT WILL INCUR A 5 POINT PENALTY. Note

that your R script file should include comments indicating the question being addressed. Your R script file should contain only the exercises/questions asked here.

▪   All variable names are shown in italics.

▪   You should discuss the interpretation of your results and how they relate to the questions you are asked.

▪   You may assume the methods you have used (e.g., linear regression) are understood by the reader and do not need definitions, but you do need to say which techniques you have used and why.

▪   As this is an assessed piece of work, you may not email/ask the module tutors questions about the essay questions.

▪   10 points will be awarded for presentation.

▪   This assessment is out of 100 marks and will count towards 50% of the term 1 mark.

The data file is natsal_3_teaching.dta. You can copy this file in the usual way from Moodle once you have agreed to the conditions of use. The variables you might use include:

Variable name             Variable description/label                                                              

Survey weight for Natsal 2010 (core and boost samples) Scrambled psu identifier

Strata

Respondent's age at interview, years

Respondent's sex

Body Mass Index

Screen positive for current depression

Relationship status

Age of respondent at birth of 1st child

Respondent had their first child before the age of 18 years Sexually conservative attitudes scale

Highest educational qualification

Household income per year (incl benefits, pensions etc) Respondent's NSSEC code (5 groups)

Government Office Region

A brief user guide can be foundhere.

DOs and DON’Ts

-     DON’T include raw variable names in your text or tables

-     DON’T use too many decimal places, but be consistent

-     DON’T include unedited R output in the main text of your essay or you will lose marks

-     DO make sure tables and figures have titles and are referenced in the text

-     DO make sure your tables and figures can be understood without reading the text

-     DO make sure you have given a clear enough description of what you have done so that the reader can reproduce any numbers/results that you present

-     DO be careful how you use the terms ‘significant’ and ‘correlation’ because they have specific meanings in social statistics.

PART A:  Multiple Linear Regression (60 Points)

This question uses the natsal_3_teaching.dta dataset. The National Childcare Trust were impressed with your reports last year and have recommissioned you to write a short report on age at first birth of women who have completed childbearing . They suggest that you  subset your analysis to those aged 45 years  and over. They  are interested to find out more about the cause of the relationship between age at first birth and the sexually conservative attitudes scale in women. They have asked you to model this relationship. You may choose to add additional antecedent explanatory variables to your model that may explain the relationship between sexually conservative attitudes and age at first birth in women. You should report any decisions you take to adjust for survey non-response in your data and to take account of any complex survey design. These decisions should form an introduction that also includes a brief description of your dataset, selection of a complete case study sample, sample characteristics. Briefly explain any limitations to your analysis in a concluding section that also summarises your main substantive findings.