Idea Density in Exam Performance

Overview

Text can be written in multiple ways to mean the same thing; changing how a text is written can make it easier or harder to understand. How many concepts or ideas there are in a text, divided by the total number of words, is one possible way to determine how easy or hard it is to understand. This ratio is called idea density (ID). Varying ID has been shown to affect the speed at which a reader understands; it impacts certain people more than others, such as second language speakers. This effect may be of particular importance in an exam, where understanding a question in a limited time is key. In the UK, pharmacy students must undertake an exam set by the General Pharmaceutical Council (GPhC) to be registered as pharmacists. The exam involves pharmaceutical calculations and shows variable pass rates. This study aims to evaluate the impact of reducing ID in a pharmaceutical calculation test and will be conducted in 14 schools of pharmacy in the UK. All participants will take a GPhC style test. Then, participants will be divided into two groups of equal size; one group will undertake a second test with the same ID as the first, while the second group will undertake a test with a lower ID. Finally, the investigators will compare the second test scores between the two groups as cohorts and question by question, evaluating whether lowering ID has increased students' scores. If and effect is seen, ensuring that questions are written with a controlled ID may help ensure we are examining more fairly and allowing students with the requisite knowledge to pass.

Pharmaceutical Council must be undertaken. It involves pharmaceutical calculations and shows variable pass rates. Linguistic factors, such as idea density, affect and predict comprehension time. This trial will evaluate the effect of lowering question idea density on attainment in a pharmaceutical calculations exam aligned to that of the General Pharmaceutical Council. Methods: This is a single-blind, parallel 2-arm multicentre randomised controlled trial conducted in fourteen Universities across the United Kingdom. A 1:1 randomisation and a sample size of 188 pharmacy students will be sufficient to detect a 1-point difference in the mean scores between the intervention and control group during a pharmacy calculation test with two-tails, 80% power and 5% significance level. Each school will recruit a minimum of 14/15 students. Participants will sit two 12-question pharmaceutical calculation tests. All participants will take the same baseline test; then, will be randomised and undertake a second test 2-week after, with standard idea density for the control group and lower idea density for the intervention. Primary outcome: the scores obtained by the participants undertaking the second calculation test 2-week after the baseline. Secondary outcomes: percentage of students achieving a pass during the second test; effect of demographic characteristics (first or not-first English language speakers, age, ethnicity, year of study, specific learning disability) on participants' attainment when lowering idea density. Conclusion: Results could inform the development of new standards in the pharmaceutical calculations exam.