Dependent t test

Card Set Information

Author:
Trekofstarsx
ID:
298422
Filename:
Dependent t test
Updated:
2015-03-15 00:26:06
Tags:
test
Folders:
t test
Description:
Dependent sample t test
Show Answers:

Home > Flashcards > Print Preview

The flashcards below were created by user Trekofstarsx on FreezingBlue Flashcards. What would you like to do?


  1. Dependent samples t
    -the dependent variable is measured two or more times for each individual in a single sample
  2. What are dependent samples?
    -Scores in one group are related to scores in another group
  3. How do we get dependent samples
    1.
    2.
    -When you measure the same people more than once (repeated measures)

    -Matched samples (someone in sample A is paired with someone in sample B) and they are identical on some measurements
  4. Repeated Measures design

    1.
    2.
    • -Each subject is measured more than once in your study
    • -You have one sample of people, but they provide two or more samples of data
  5. Matched Samples Design
    • -Two or more samples of people that provide one sample of data each
    • -People matched to each other in pairs
    • -Treated as though they are one sample
    • -Often used in quasi-experiments when you can't randomly assign people to conditions
  6. All statistics are done with
    Difference scores (d=x2-x1)
  7. What test do we use?
    1 sample t
  8. Mean difference formula
    see chart
  9. H0
    uD=0
  10. Standard error dependent t test
    See chart
  11. Sum of squares D
    Variance D
    see chart
  12. Cohen's D for mean difference
    -Mean difference in standard deviation units
  13. Benefits of Repeated measures

    1.
    2.
    3.
    • -Reduces error due to individual differences
    • -Reduces number of participants
    • -Can observe changes over time (longitudinal studies)
  14. Carryover effects
    -The first treatment or measurement influences scores on the second measurement
  15. Progressive error
    -Passage of time alone can produce change
  16. Solutions to carryover and progressive error

    1.
    2.
    • -Counterbalance order of treatment
    • -Choose a different research design
  17. Assumptions of related-samples t-test

    1.
    2.
    • Observations within each treatment are independent
    • Population distribution of difference scores must be approximately normal

What would you like to do?

Home > Flashcards > Print Preview