Why must a study be conducted under controlled conditions?
So we can suggest that the effect has been found due to the identified variable, not other factors that weren’t tested for
What happens to an IV in a study?
The independent variable is the variable that gets manipulated
You are able to test a specific hypothesis by manipulating an IV
You can measure the effect of the manipulation using the DV
The difference between reliability and validity?
Reliability is the consistency of a measure and the validity is that the study is examining what it aims to examine (measures what is is meant to measure)
What are some strengths of an experimental design?
- They are scientific, so they follow a standardised procedure, meaning it can be replicated
- experiments enable a researcher to control for extraneous variables
- Experiments produce quantitative data that can be subjected to statistical analysis, which ensures that results of different groups of participants ca be meaningfully compared
What are some weaknesses of an experiment design
- often low in ecological validity (meaning it is low in generalisability across places and people) as they are generally controlled in situations removed from real life (or a perfectly conducted CBT trial)
- sample sizes are often small
- can case stress to participants (ethics)
- results are generally reduced to numerical data, which can be limiting
What is a strength and weakness of an independent measures design
Even with random allocation it is not possible to control pp variables effectively because sample size is normally small
- pp variables are related to individual characteristics of each pp that may impact how they respond
+ no order effects, demand characteristic are reduced as they participate only once
What is a strength and weakness of a repeated measures design?
- order effects need to be controlled, greater likelihood of: demand characteristics, order effects practice effects and boredom effect
+ removes PP variables, as each op is tested in all conditions
Examples of self-reports
Interviews / surveys
Pros and cons of self reports
+ allows collections of large amounts of data quickly
+ usually standardised and easily produced
+ can produce both numerical and qualitative data
- subjective method of data collection
- interviewing can be time consuming
- thematic analysis etc can be more bias and subjective / ethnocentrism
- demand characteristics
Pros and cons of qualitative data
+ Not limited by preselected categories
+ Can be analysed in a bottom up way
+ Rich and meaningful data that can have more impact / voice of the individual is heard / implications for future research + practice / can help enhance proficiency of services
- Can be difficult to analyse sometimes
- takes a long time
- how we interpret the data can be bias
- can be difficult to draw meaningful conclusions from
Pros and cons of quantitative data
+ allows for statistical analysis
+ we can suggest a difference / correlation / demonstrate statistically meaningful relationships between variables
+ cause and effect can be measured
+ relationships between variables can be analysed
- can impose limited choices and by imposing categories we can miss out on potentially important aspects of behaviours or feelings (nod to when completing ROMs with a YP, doing it with them, as a rapport building point but further to gain richer more detailed info that quan tests often don’t allow for)
What are some positives and negatives of a correlation study?
+ identifies relationships and trends
+ data is often from real world settings (increasing ecological validity)
+ findings often provide a strong foundation for future more intensive experimental research (such as a regression analysis)
+ researchers can study variables that are often difficult to control
+ efficiency and cost effectiveness (completed quickly)
- it cannot prove causation (we can only state if there is a relationship between variables)
- confounding variable problem - an unseen / unknown variable might be driving the relationship between the two
- susceptibility to data bias (the quality of the data is heavily reliant on self report and surveys which can produce bias!)
Design a study and Name the IV(s) and DV(s) for the following: anxiety, sleep, social media use
DV - anxiety
IV - sleep
IV - social media use
There is a relationship between sleep, social media use and an individuals anxiety score
When might you use a standard multiple regression analysis?
- To asses how much variance in a dependent variable can be predicted by a set of independent variables (eg how much variance in work performance ca be predicted by conscientiousness and agreeableness)
- To determine the amount of unique variance an independent variable predicts in the dependent variable (eg does conscientiousness or agreeableness predict more unique variance in performance at work?)
Why use a multiple regression analysis?
It allows your to predict scores on a dependent variable
When might you use a hierarchical multiple regression analysis?
To assess whether independent variables are able to Pret g the dependent carbs or whilst controlling for extraneous variables (eg does conscientiousness and agreeableness still predict a signal amount of variance in performance if we control for intelligence?)
To assess whether adding additional IV’s to a model improves the ability to predict the DV
