Latent Factor Analysis: 3 Dec 2014 – covers latent variables and factor analysis at an introductory and intermediate level. A latent variable is something invisible (such as a concept, an attitude, or an illness) that cannot be measured directly that has been measured using a set of related observed indicators. Factor analysis is one way to derive a single factor from a set of variables, and is thus called a data reduction method. Other data reduction methods include principal components analysis, which is very closely related to factor analysis, and multiple correspondence analysis. We will focus on confirmatory factor analysis, but talk a bit about the differences with exploratory factor analysis. The course is suitable both for primary-data collection researchers (who may need to write a suitable questionnaire), and for those who want to analyse secondary data sets.
An Introduction to Computational Social Science using Big Data: 5 Dec 2014 – every day we interact with countless technological systems, which support our communication, our transport, our shopping activities, and much more. Through these interactions, we are generating increasing volumes of “big data” and there is scope for measuring human behaviour, captured in a natural setting at an unprecedented speed and scale. Such data constitute a new opportunity for social science research. To make maximum use of these datasets, researchers must possess a combination of programming skills and statistical analysis skills, alongside subject specific knowledge. Participants should be interested in learning basic computer programming and statistical analysis skills, and applying these to datasets.
Forecasting Methods and Models: 10 Dec 2014 – aims to cover the commonly used techniques to forecast demand in public services and in business. The emphasis of this course is on the practical application of forecasting techniques rather than on theoretical content. Alongside quantitative techniques we will also cover the more qualitative and judgement based aspects of forecasting.
Social Media Data Analysis: 11 Dec 2014 – designed for those conducting research on the web or for those conducting a non-web research that wish to assess how their topic is reflected in the social web
Statistics for Small Samples: 16 Dec 2014 – covers bivariate statistical tests for a variety of situation. The basic material of t-test is enriched by adding methods for the comparison of the distribution (or a mean) in cases where the variable is not normally distributed. The word ‘parametric’ is covered so that nonparametric approaches generally can be taken up confidently. The course is at an intermediate level, in the following sense: we hope all participants will read about (or review) the Chi Squared test in preparation. We then move to the small-samples variant of this test, and other tests for varying situations. Time permitting, we also cover the Kolmogorov-Smirnov 2-sample test, the Kruskal-Wallis test, the median test and Spearman’s coefficients, and one other test statistic. The participants learn to critically assess the validity of claims to statistical inference (from sample to population) in small-N and medium-N situations. Helpful examples are provided in the form of primary survey data from the Young Lives Programme data on care-givers of children born in the year 2000 in four countries, and from a large employment dataset.