Research Methods L2 Research design Repeated measures- you yourself having different treatment repeated on yourself, one of the problems with that is that you can overcome it, and familiarise yourself with the pressure. But what you can do is randomise it. Music & Endurance study during running- People all respond differently to individual preferences Ignoramus- 'we don't know' - The things that we think we know, could be proven wrong as we gain more knowledge - No concept, idea or theory is sacred and beyond challenge When designing a research question, it needs to have originality, and applied to a specific population. Who will your findings be useful to? What is the benefit of home physical activity barriers? - Travel - Expenditure - Socio-cultural differences Limitations: Who would actually stick to it? Method or methodology - Using a particular research design for you research enquiry - Includes: research question, objectives, reason for choosing the specific method - Why? How? Method: Research equipment used to measure and analyse data Who? What? Where? When? How? Objectives- methods used to answer aim, the way we do it Aim- What we want to achieve Triangulate - Taking two or more methods or measures. Quantitative research Aim- to determine how one thing (an independent variable e.g. age, gender) affects another (a dependant variable) in your group. Designs- 1. Descriptive (subjects usually measured once) Establishes only relationships between variables e.g. tracking physical activity levels of second year uni students. Physical activity monitors, what you've eaten and slept through the week. Wide variability. - Can be useful, however need to think about controlling group/cohort Sample= >100-10000 participants to estimate of relationship between variables 2. Experimental (subjects measured before and after a treatment) Interventional study. Establishes causality- can't be proven, but can allow us to see potential causes. An experiment, especially a crossover, may need only tens of subjects. E.g. Effect of redbull being drank during a lecture. Descriptive Designs Observational studies (longitudinal): Cohort studies, case control studies, cross-sectional studies and case series. Cohort series: Large samples
Case series: take a specific group of people and you just observe them. Experimental designs Intervention studies: Pre post single group- giving someone something, look at before and after Pre-post crossover- Pre-post parallel groups- Two groups running side by side. Each group doing a different thing, still look at before and after. Post-only cross-over/parallel- Take measurement Pre-post single group Weakest design- any change after treatments could be coincidental No control group- so can only look at interventional effect. How to improve? - Blind participants - Limit 'random' changes by: series of baseline trials, repeating the treatment with the same group after washout. - Take 3 times, allows you to check variations, and gives an average result. Pre-post crossover- repeated measures design Best design to estimate treatment effect on participants Why ?- all participants receive all treatements - you can estimate individual responses Sample size- half that for parallel groups, but twice as many trials - Save on participants, the same people