Application Flashcards Preview

Stats > Application > Flashcards

Flashcards in Application Deck (11)
Loading flashcards...

What are the steps to conducting research?

1. Devise a research question (based off previous research)
2. Design a study to test the question
3. Choose the stimuli, e.g. questionnaire
4. Apply for ethical approval
5.Run the study and collect data
6. Analyse the data and see what you found out
7. Write up your study as a research report


What is continuous variables?

Scores exist on a continuum e.g. IQ, reaction time, accuracy


What are categorical variables? (binary)

2 options
e.g. sex, verdict, handedness


What is a categorical variable? (3+ categories)

IMPORTANT: no linear order
e.g. drug type, hair colour, degree


What variables do you use in correlational research?

-Both must be continuous
-No groups


What needs to be considered when running a linear regression?

-Outcome: must be a continuous score
-Predictors: must be continuous or binary
-Can't have more than 3+ categories predicting


What is an effect size?

-Is significance enough?
-Effect size is unaffected by sample size
-ES is typically standardised, not influenced by scale of measurement and comparable across studies (meta analysis)


What effect sizes are there?

small, medium and large


What is power analysis?

-How many participants are enough?
-What is the chance of detecting an effect? (% chance)


How do you calculate a sample size?

-Size of the effect to be detected (effect size)
-Desired alpha level p = .050
-Desired level of power - typically .80 (80& chance of detecting an effect size if it is there)


When should power be used?

Only at the planning and design stages of an experiment