Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
Research Design
exam 3
| Question | Answer |
|---|---|
| Types of within group designs and why | natural- siblings, t-test; natural sets- ANOVA; matched- matched on variable, matched sets- ANOVA; repeated measures- time 1, time 2 |
| T-tests (independent vs. dependent) | independent- btw 2 groups, dependent- depends, 1 group measured 2x or matched pairs measured once (natural or matched) |
| One-way ANOVA, when to use, how to read | 2 versions: randomized- btw groups, repeated measures- within groups; 3 times |
| Assumptions of repeated measures ANOVA | interval/ratio, homogenous variance, normally distributed, data correlated |
| Benefit of repeated measures | controls individual differences |
| Eta-squared | (ANOVA)- effect size |
| When to use post hoc tests | when significant for ANOVA |
| Advantages of correlated t-tests | reduce individual error |
| confounds | uncontrolled extraneous variables |
| how to control | Latin square, counterbalance |
| internal validity | if IV had an effect on the DV |
| external validity | generalizability |
| Assumptions of a randomized ANOVA | observations are independent, interval/ratio, homogenous variance, normally distributed |
| Between group variance= | systematic (effect of IV) +error (us being human) |
| IV affects between group variance (big vs. small) | F is big for bigger effect of IV, small is close to 1 |
| Figure df for randomized ANOVA= | (k-1), (n-k), correlated t-test= n-1 |
| F ratio= | between/ within variance |
| Large within group variance= | difficulty in telling if the IV has an effect |
| Systematic variance= | effects of the IV & includes confounds |
| Benefits of a multi-group design | (compare treatment types- humanistic, psychodynamic, CBT, control & 2 experimental, or placebo) |
| Bonferroni | minimize type 1 |
| Between vs within group design | btw- easier to create, within- harder to create but have more power of significance |
| When to accept or reject the null | when p< .05, when the observed value is higher than the critical value |
| how to increase power | significant difference, small variability, big sample |
| equivalent groups | matched subjects strength, create equivalent groups, more power (match age, intelligence, gender) |
| Carry-over effects, order effects-within | threat to within group designs |
| Cohen’s d | t-test effect size |
| effect of sample variance on t | as variance increases, t value decreases; too much chance, not enough btw group difference |
| College sophomore problem | too much studies on college students, data skewed |
| types of replication | exact- do it the same way, conceptual- a new population, change more things, measures, or methods; systematic- changing 1 thing (1 level of the IV or scale or could add a layer) |
| Floor effect | cutoff |
| Ceiling effect | no one scores higher |
| single-blind | participant doesn’t know who gets the treatment but the researcher does |
| double-blind | both don't know |
| Diffusion of treatment | 1 person talks about it & it diffuses, Christmas example; word spreads |
| Threats to internal validity | affects the IV |
| Threats to internal validity: | Mortality- loose participants or die; history- something in the world is going on, ex. Covid; instrumentation- use a broken MRI, wrong scale, enter things wrong/bias; maturation- people change |
| Regression to the mean | if low, will go back to the mean, score above, you’ll regress to the mean |
| Fatigue, practice effects | done more than once, will score be higher; threats to internal validity |
| control group | baseline |
| experimental group | treatment |
Created by:
Finch35i