Undergraduate Course: Doing Social Research with Statistics (SSPS08007)
Course Outline
| School | School of Social and Political Science |
College | College of Humanities and Social Science |
| Credit level (Normal year taken) | SCQF Level 8 (Year 2 Undergraduate) |
Availability | Not available to visiting students |
| SCQF Credits | 20 |
ECTS Credits | 10 |
| Summary | Providing Intermediate level Statistical Tools for Students in the Quantitative Methods Degrees
This course is designed to allow students in the with Quantitative Methods degree programmes in SPS to move beyond basic statistical techniques into intermediate-level techniques, which will later enable them to learn advanced techniques. Therefore, it aims to lay the foundations for advanced techniques: Considering the ways in which secondary data is produced; Moving beyond linear regression to models based on log-odds to predict categorical results; data reduction; analysis of variance between groups. A well trained analyst should have acquired skills uing a variety of software packages that are commonly used in social research, and as such this course introduces Stata and R, in addition to SPSS. The course is aimed at students who also study Sociology, Social Policy, Politics, and International Relations. As such, it will contain examples and applications relevant for all these disciplines.
|
| Course description |
The course starts with considering issues of quantitative data collection, moves to regression models based on a logarithmic transformation to predict odds rather than a specific category, considers methods which go beyond regression models, and finally introduces two prominent software packages in addition to SPSS.
In the first part (weeks 1-2), the course will discuss the two most prevalent ways to produce datasets for social statistics: Surveys which look at a sample to examine a larger population, and digital sources of data such as administrative and transactional records, and the web (http://www.esrc.ac.uk/research/research-methods/dsr.aspx). Both the ways in which the data are produced and the ways to access and use data will be discussed.
The second part (weeks 3-5) of the course teaches three techniques to produce regression models with categorical variables as dependent variables. These three techniques cover the full range of categorical variables: binary, nominal and ordinal. Using these techniques requires a transition from predicting a specific outcome in linear regression, to producing odds a ratio of probabilities of belonging to different categories.
The third part (weeks 6-8) of the course looks at multivariate analysis which goes beyond the regression models usually used in the social sciences. First, it looks at ANOVA (and ANCOVA), a method commonly used in social sciences which conduct experiments (such as psychology and linguistics), but is also useful for a range of other examples. Second, the course will look at two data reduction techniques Exploratory factor analysis (or principal component analysis) which aims to find an underlying concept within different variables, and structural equation modelling, which maps relationships between variables in a spatial way.
In the fourth and final part (weeks 9-10) the course will introduce two further software applications: Stata and R. Since at this point the students will already be familiar with a wide range of both basic and intermediate techniques, they might wish to expand their analysis beyond the limits of SPSS. Both Stata and R offer a wider range of techniques and a larger degree of control over the way the analysis is carried through compared to SPSS, and the students benefit from having a range of software to use in their analysis. Throughout the course, practice in presenting the results of complex statistical analysis will be provided, so that students thinking about clarity of presentation as they are learning new techniques. Week 11: Conclusion & revision.
Part 1: Production of Quantitative Data
Week 1: Survey data
Week 2: Digital social research
Part 2: Regressions with Categorical Dependent Variables
Week 3: Binary logistic regression
Week 4: Nominal regression
Week 5: Ordinal regression
Part 3: Beyond Regression Models
Week 6: Anova
Week 7: Exploratory factor analysis
Week 8: Structural equation modelling
Part 4: Using Alternative Software to SPSS
Week 9: Stata
Week 10: R
Week 11: Conclusion & Revision
|
Course Delivery Information
|
| Academic year 2015/16, Not available to visiting students (SS1)
|
Quota: None |
| Course Start |
Semester 2 |
Timetable |
Timetable |
| Learning and Teaching activities (Further Info) |
Total Hours:
200
(
Lecture Hours 11,
Seminar/Tutorial Hours 22,
Programme Level Learning and Teaching Hours 4,
Directed Learning and Independent Learning Hours
163 )
|
| Assessment (Further Info) |
Written Exam
100 %,
Coursework
0 %,
Practical Exam
0 %
|
| Additional Information (Assessment) |
40% mid-term exam (comprised of multiple-choice questions)as a formative feedback event.
60% take home exam (students will conduct a series of analysis tasks and report them).
|
| Feedback |
Not entered |
| No Exam Information |
Learning Outcomes
On completion of this course, the student will be able to:
- Make effective use of regression models for categorical dependent variables, including binomial logistic regression, multinomial logistic regression, and ordinal logistic regression.
- Make effective use of a regression model for limited dependent variables¿namely, tobit regression.
- Use the statistical analysis software package Stata to estimate regression models for categorical and limited dependent variables.
- Interpret estimates from regression models for categorical and limited dependent variables using different approaches.
- Understand alternative specifications of regression models for categorical and limited dependent variables as a latent variable, nonlinear probability, and linear model.
|
Reading List
The course will suggest that students use both books and on-line resources to facilitate their learning.
Argyrous G (2005). Statistics for research: with a guide to SPSS (2nd edn), Sage,
London.
Pampel, FC (2000). Logistic regression: a primer, Quantitative Applications in the
Social Sciences, 132, Sage University Papers, Sage, London.
de Vaus, D (2013) Surveys in Social Research, 6th ed., London: Routledge.
|
Additional Information
| Graduate Attributes and Skills |
Not entered |
| Keywords | Not entered |
Contacts
| Course organiser | Dr Alexander Janus
Tel: (0131 6)51 3965
Email: Alex.Janus@ed.ac.uk |
Course secretary | Mr Daniel Jackson
Tel: (0131 6)50 3932
Email: Daniel.Jackson@ed.ac.uk |
|
© Copyright 2015 The University of Edinburgh - 18 January 2016 4:53 am
|
University Homepage