Learning Outcomes:
On completion of this module the student will:
- understand data and the implications of type of data and level of measurement for subsequent analyses;
- compute and interpret descriptive statistics appropriate to the data;
- understand sampling variation, significance testing and confidence intervals;
- compute and interpret confidence intervals for means and proportions;
- generate research hypotheses and know how to test them;
- know the assumptions underlying statistical techniques;
- choose and carry out appropriate statistical techniques for two-group comparisons;
- carry out appropriate statistical techniques for independent and paired comparisons;
- derive, present and interpret p values;
- understand, compute and interpret correlation coefficients and simple linear regression models;
- understand and interpret univariate survival methods;
- use statistical software to carry out descriptive and comparative analyses;
- have a broad overview of multivariable analysis.
Note that at the module coordinators discretion a viva voce may be used as an oral assessment, for some students.
Indicative Module Content:
- Data, types of variables, levels of measurement, distributions of data;
- Descriptive statistics, measures of central tendency, measures of dispersion and position;
- Methods used to summarise and present descriptive data;
- Statistical distributions; Basic probability theory;
- Sampling variation; Confidence interval estimation;
- Standard errors for means and proportions. Confidence intervals for means & proportions;
- Hypothesis testing, statistical significance, Type 1 and Type II errors;
- Assumptions underlying statistical tests. Parametric and nonparametric techniques;
- Comparisons between groups; Statistical techniques appropriate to data and assumptions;
- Significance tests for differences in means (z test, t tests); CI's for differences in means;
- Significance testing for differences in proportions (chi sq tests, McNemar Test); CI's for differences in proportions;
- Non parametric tests for differences in distributions / medians;
- Correlation, simple linear regression;
- Use of online statistical software and SPSS;
- Introduction to use of multivariable methods.