LI0033: Statistics and Experimental Design

LI0033 Term 1

Statistics and Experimental Design

Lecturer	Dr. Ellen Gurman Bard ellen@ling.ed.ac.uk, AFB-322 office hours: Tues 10:30-11:30, Wed 15:30-16:30
Tutor	James Horlock horlock@cstr.ed.ac.uk

page created on October 3, 2000

Aims

This course should teach you how to approach the design and analysis of scientific studies of human behaviour. It will impart general principles of experimental research, issues of importance in the design of studies, ways of thinking about and treating the results. The main goal of the scientific enterprise is to know more with more certainty and less effort. The course should have direct impact on your own project work, saving you much time and many lines of prose which you could otherwise waste in trying to understand what your own data mean.

Methods

The course will include lectures (Wednesday of every week; Friday of week 1) and tutorials (Friday 14:00-15:00 or 15:00-16:00). Lectures describe principles and methods. Tutorials show you how to use a statistical package to apply methods. You may use the PCs in the AFB-B12 to supplement the weekly tutorials. Tutorial assignments will always include a problem to solve on your own. In addition, you will be assigned design problems which involve thinking rather than calculation. These will help you think like a critical experimenter. They should improve your judgment as a consumer of experimental research and help you design good studies of your own.

Assessment

Honours students will be assessed by an end-of-term project and an end-of-year exam. MSc students will submit an end-of-term project as part of their December exam. These projects will comprise a series of questions about a new data set. All students intending to do Honours or MSc dissertations will be expected to propose work which is satisfactory with respect to its design and susceptible to rigorous analysis.

Reading

I will try to have copies of the assigned texts put on reserve, but note that the library is re-cataloguing and this may take some time.

Assigned texts:

Coolican, Hugh (1999). Research methods and statistics in psychology (3rd edition) London, Hodder & Stoughton.

Greene, Judith and D'Oliveira, Manuela (1982). Learning to use statistical tests in psychology: a student's guide. Second edition. Milton Keynes, Open University Press.

Kinnear, Paul R. and Gray, Colin D. (1999 or 2000). SPSS for Windows made simple . Hove: Psychology Press.

We will keep a couple of copies of this in the lab.

Useful for reference:

Anshen, F. (1978). Statistics for linguists. Rowley, MA:

Aron, A. & Aron, E. (1999). Statistics for psychology. Upper Saddle River, NJ: Prentice Hall.

Butler, C. (1985). Statistics in linguistics. Oxford: Blackwell.

Howell, David C. (1997). Statistical methods for psychology(4th edition) Belmont: Duxbury Press.

Woods, A. Fletcher, P. & Hughes, A. (1986) Statistics and language studies.

Easy, funny or lovable for one reason or another:

Cohen, J and Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale: N.J. Lawrence Erlbaum.

Gonick, L., & Smith, W. (1993). The cartoon guide to statistics. New York: HarperPerennial.
Dancey, C., & Ready, J. (1999). Statistics without maths for psychology. Prentice Hall.

Siegel, S (1988). Nonparametric statistics for the behavioral sciences. London : McGraw-Hill,

Two classic articles on experiments using linguistic stimuli:

Clark, H. (1973). The language-as-fixed effect fallacy: a critique of language statistics in psychological research. Journal of Verbal Learning and Verbal Behavior, 12, 335-359
Raaijmakers J., Schrijnemakers J., & Gremmen F. (1999). How to deal with ''The language-as-fixed-effect fallacy'': Common misconceptions and alternative solutions. Journal of Memory and Language, 41(3), 416-426

Interesting statistics links

Follow the link below to other very useful sites with statistics information, often themselves containing links to further sites.

External Statistics Links

Schedule

C = Coolican; G&D = Greene & D'Oliveira

WK		Wednesday 14:00-15:00 , Room AFB-B9			Friday 14:00 - 15:00 or 15:00-16:00 Room AFB-B12 (except wk 1)
		Topic	Reading
1	[11.x] Introduction: Science as a Research Method			C, ch. 1 (WE) C, ch. 2-4 (FR)		(AFB-B9) Experimental Design - assuring you can know more from less
2	[18.x] Measurement Scales & Descriptive Statistics			C, ch. 10		Lab 1: Descriptive Statistics & Data Exploration
3	[25.x] Relationships between two variables: Ordinal vs interval measures of correlation			G&D, ch. 8; C ch. 15 pp 346-59		Lab 2: Correlation
4	[1.xi] Nominal Data: from description to statistical inference: Binomial/Sign test, Chi-Square			C, ch. 11-12		Lab 3: Binomial/signTest, Chi-Square
5	[8.xi] Ordinal Data: Wilcoxon, Friedman, Mann-Whitney, Kruskal-Wallis			G&D, ch. 5-7		Lab 4: Wilcoxon, Friedman, Mann-Whitney, Kruskal-Wallis
6	[15.xi] Interval/ratio data: z-scores again, t-tests			G&D, ch. 10-12.		Lab 5: t-tests
7	[22.xi] Interval/ratio data: ANOVA - basic concepts			G&D, ch. 13-14		Lab 6: ANOVA part 1
8	[29.xi] Interval/ratio data: ANOVA - within subjects designs: with bells and whistles			G&D, ch. 15; C, ch. 19		Lab 7: ANOVA part 2
9	[6.xii] Interval/ratio data: Regression and Multiple Regression.			C, ch.. 15		Lab 8: Regression & Multiple Regression.
10	[13.xii] Sampling and people: special designs and considerations			C, ch. 3, Papers by Clark and Raaijmakers et al		Lab 9: Test or extend your SPSS skills.