|Contents:||The aim of the course
Prior statistical / methodological knowledge and skills
- Get acquainted with and get understanding of several important multivariate statistical techniques . Successively the following subjects will be discussed:
- fundamental statistical concepts/elementary probability topics
- correlation and regression analysis|
- analysis of variance (one-way ANOVA, multi-way ANOVA, ANCOVA, repeated measures, multivariate ANOVA )
- discriminant analysis
- factor analysis (principal component analysis)
- cluster analysis
- multidimensional scaling/correspondence analysis
- Get acquainted with and get understanding of several advanced and current methodological topics, using a topic list 'capita selecta research methods and methodology' : varying from classical topics such as (experimental) research designs, validity, reliability, generalisability, causality and confounding, to more current topics such as meta-analyses, multi level research, grounded theory and action research.
- Get understanding of the interplay between statistics and methodology on the basis of the lectures on statistics and the study of the named selection list. The student must be able to design and carry out a small research project : to do a problem analysis, to formulate a research objective, to formulate an appropriate problem statement and research questions, to conduct an advanced data analysis using SPSS, to interpret the results, to identify and avoid possible methodological pitfalls and finally to indicate how in further research the problem statement can be investigated in a better way.
Some general knowledge of the following topics is required prior to the course. See below. Some of these topics will re-appear again in ARM, they will be reviewed thoroughly or even be treated in more depth in Kachigan.
Students with no introductory knowledge of the following statistical topics are strongly advised not to take the ARM-course, but to take a more introductory course first.
Elementary inferential statistics:
- measures of central tendency (mean, median, modus), measures of disperion (range, variance, standard deviation, IQR), percentile values, kurtosis and skewness, frequency distributions, relative frequency and cumulative frequency, z-scores, empirical and theoretical distributions, the normal distribution, visualisation techniques (barcharts, piecharts, histograms, boxplots, scatterplots)
Statistical tests for two of more variables:
- sample versus population, sampling distributions, standard errors, central limit theorem, z-test, one sample t-test, degrees of freedom, the logic of hypothesis testing (significance level, p-value, one-sided and two sided testing, type 1 and type 2 errors) parameter estimation (confidence intervals), parametric versus non-parametric statistics
- chi-squared analysis, correlation analysis, two-group t-test, paired t-test, one-way ANOVA, simple regression-analysis
- The student must have a working knowledge of SPSS 10.0 or higher to make datafiles, manipulate data (recode, compute, etc.), vizualize
data, perform the above mentioned statistical techniques and interpret the SPSS outcome
Elementary Research Methodology
- The research process
- Conceptualisation and measurement: levels of measurement (nominal,ordinal, interval, ratio), measurement validity (face validity, content validity, criterion validity, construct validity), reliability (alternate-forms reliability, test-retest reliability, inter-item reliability, inter-observer reliability)
- Sampling methods: probability sampling (simple random, systematic random, cluster, stratified) versus non-probability sampling (quota,snowball,purposive, availability)
- Research design (experimental versus observational, causation and confounding, internal validity versus external validity, threads to internal validity)
- Experiments (pre-experiments, quasi experiments, pure experiments)
- Basics of qualitative research (comparison with quantitative research, validity and reliability, participant observation, intensive interviewing, focus groups)
The Examination comprises three parts:
Further information will be given during the course.
- T1 ("Test 1") A written examination halfway the course, covering all topics discussed in the lectures1, 2, 3, and 4. T1 is a "closed-book-exam" only a pocket calculator is allowed
- T2 ("Test 2") A written examination at the end of the course, covering all topics discussed in the lectures 5,6,7 and 8. T2 is a "closed-book-exam" only a pocket calculator is allowed
- G (Group assignment ) Research Project including an advanced data analysis on a real-life data set and the writing of a scientific report
The student has passed the course if and only if the following inequalities hold: 0,5(T1+T2)>= 5.6 and G >= 5.6
If these conditions have been fulfilled, the final grade is computed as follows:
Final grade = 0,6 T + 0.4 G where T= 0,5 (T1+T2)
If these conditions have not been fulfilled, the final grade is computed as follows:
Final grade = minimum [ 0,6 T + 0.4 G), 5] So, if one or more of the beforementioned conditions has not been fullfilled, the highest possible grade for the entire course is a 5 , which implies that the student has not passed the course.
A re-examination is only possible for T If T <=5.6 for example, because the student does not make T1 or T2, than the only way to pass ARM is to make a re-examination TT covering all lectures. TT and T are equally weighted.