Generalizability and Credibility in psychological research and how to increase it are the topics in today’s episode of The Psychology World Podcast.
Hello, everyone I hope that you’re having a great week.
In addition, to what generalizability and credibility is in psychology research, we will be looking at how to increase generalizability and credibility in research as well.
So today’s show notes are taken from my Research in Psychology book:
Generalizability refers to the extent to which the findings can be applied to other populations and other settings that weren’t used in the study.
For example: applying the results of a drug study on teenagers to adults as well.
Credibility refers to the extent to which the findings can be trusted to reflect the reality of the world. The higher the creditable the more trusted they can be to effectively tell the truth.
Generalizing the findings:
Over this section, we’ll be looking at factors that will increase a study’s ability to generalise their findings.
Ecological validity:
This means the extent to which the findings can be applied to the real world as a result of the experimental scenario.
Studies with low ecological validity tend to use artificial scenarios and ones that don’t match the real world. Therefore, the findings cannot be applied to the real world or other populations.
Studies with high ecological validity use real-world scenarios so you can apply the findings to other situations and populations.
You can increase the ecological validity in your experiments by making your experiment as lifelike as possible.
Population validity:
Meaning the extent to which your findings can be generalised from your sample.
Studies with low population validity tend to use inappropriate samples or too small sample in
relation to their target population for a meaningful conclusion to be drawn.
For example: if I was to do a study on teenage drug addiction and I used only 20 12 and 13-year-old then I would have low population validity. As I’m only using a small age group and I’m using 12-year olds that aren’t even teenagers. Plus, I’m only using 20 of them compared to the millions of teenagers in the UK.
Whereas, studies with high population validity use a large sample size using an appropriate population to study.
Therefore, to improve the example above, I would use 500 13-17 years old.
Construct validity:
Construct validity is how well does the method for measurement measure what you want to measure.
Such as: how well does rating your anger on a scale of 1-5 help you measure anger levels.
Studies with low construct validity tend to use inappropriate measurement tools.
For example: using a focus group when you want to test people’s shopping habits after being exposed to a drug.
Studies with high construct validity use effective measurement tools.
+Such as: conducting an experiment at a shopping centre to test shopping’s habits after being exposed to a drug.
Credibility:
Now that we have looked at what makes a study generalisable, we will now look at what makes a study more credible.
Creditability also referred to as internal validity in experiments is the extent to which the experiment or study measured what it intended to.
By doing these things in your studies, you will increase your creditability as they will increase the trustworthiness of your findings.
Triangulation:
There are many different types of triangulation including:
· Method triangulation- you use more than one method in your study to give you more data to support your conclusions with.
· Researcher triangulation- comparing and combining observations made by different researchers. Increasing credibility as if a number of different people arrive at the same conclusion then it makes it more reliable.
· Theory triangulation- you use multiple theories or ideas to analyse the data. Making your findings more reliable as you have considered other theories or things that could explain your results, and this leads to a more holistic conclusion as you haven’t tried to bring the behaviour down to one cause, but you have considered multiple causes.
Reflexivity:
This is when you take your own biases into account and you consider how they could influence the results.
Leading to an increase in credibility as this allows you to consider how you will impact the experiment and adapt how the experiment will be done accordingly. Overall, you will not be influencing the experiment as much so it should be the participant personal behaviour in the results and not their influenced behaviour.
Thick descriptions:
This means that the observed behaviour should be described in a lot of detail and in context so that it makes perfect sense to an outsider who has never seen the experiment before.
Making the study more reliable as it could mean that the study can be accurately repeated to further support the findings if the follow up get similar or the same results.
I hope that you’ve enjoyed today’s episode and if you want to know more about Research in Psychology then check out my Research in Psychology book. Available for FREE at your local library and available in eBook and audiobook on all major online stores.
Have a great day everyone!