WJEC A Level Psychology Experimental Design, Levels of Measurement and Descriptive Statistics Exam Questions

About WJEC A Level Psychology Experimental Design, Levels of Measurement and Descriptive Statistics

This area of Unit 2 Section B covers how participants are allocated to conditions and how data are measured, summarised and displayed. You need the three experimental designs with their strengths, weaknesses and controls, the four levels of measurement, the graphs in the specification, and measures of central tendency and dispersion. These are tested through short calculations and interpretation, often with a table of data in the novel scenario question. Examiners expect correct choices, such as a bar chart for categories or the median for skewed data, with a clear reason, and accurate arithmetic with working shown.

What you need to know

• Independent groups Different participants in each condition. No order effects, but participant variables may differ between groups, so random allocation is used.
• Repeated measures The same participants do every condition. Participant variables are controlled, but order effects such as practice and fatigue occur, so counterbalancing is used.
• Matched pairs Participants are matched on relevant variables such as age or IQ, then one of each pair goes into each condition. Reduces participant variables without order effects, but matching is difficult.
• Levels of measurement Nominal: categories. Ordinal: ranked data with unequal intervals. Interval: equal intervals on a scale. Ratio: equal intervals with a true zero.
• Graphs and tables Frequency tables summarise raw data. Bar charts show discrete categories, histograms show continuous data, line graphs show change, pie charts show proportions and scatter diagrams show correlations.
• Descriptive statistics Mean, median and mode measure central tendency; the median is better than the mean when there are extreme scores. Range and standard deviation measure dispersion; standard deviation shows spread around the mean.

Common mistakes in Experimental Design, Levels of Measurement and Descriptive Statistics

• Saying repeated measures removes all extraneous variables. It controls participant variables but introduces order effects.
• Drawing bars touching on a bar chart for categories. Bar charts for categories need gaps; histograms are continuous.
• Using the mean with extreme scores without comment. Outliers distort the mean, so the median may represent the data better.
• Confusing interval and ratio data. Ratio data have a true zero, such as time; interval data do not, such as temperature in Celsius.
• Leaving out working in calculations. Show each step so method marks can be awarded.

Try three Experimental Design, Levels of Measurement and Descriptive Statistics exam-style questions