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
Ad. Research Design
Exam 2
| Question | Answer |
|---|---|
| how to evaluate qualitative results: | transfer (equivalent to generalizability), credible, dependable (reliable), confirm (replication) |
| type 1 (how to control, costs) | accept the alternative when the null is true, control- set p as less than 0.05 |
| type 2 error (how to control, costs) | accept the null when the alternative is true, large sample size, strong IV |
| when to run a t-test | compare means, parametric, inferential |
| when to run a z test | single sample to known population where variance is known (know everyone’s STAAR scores) |
| assumptions of t tests | symmetrical, 30ish, pop. Mean is known but variance is unknown |
| degrees of freedom (def, how to calculate) | n-1, ability of a number to vary (wiggle room for someone to screw up) for error |
| 1 tailed vs 2 tailed (both null & alternative hypothesis) | 1 tailed is directional, 2 tailed is nondirectional |
| how to increase power | designate a 1 tailed test (directional), large sample |
| criterion validity | comparing to outside standard, use 1 score to predict another score (predictive & concurrent) |
| construct validity | theoretical, hardest 1 to prove |
| face validity | looks valid on a surface level |
| content validity | each item measures what it is supposed to measure |
| control | find out if IV affects the DV |
| science is | self-correcting |
| assumptions of z tests | symmetrical, over 30, pop. M/SD is known- standard deviation & z are connected, mean of 0 |
| SEM is | SD of sample |
| SEM is based on | Central Limit Theorem, smaller than SD |
| sample vs population | sample comes from population, sample distribution is a sample of a random sample of a population (think M&Ms) |
| parametric vs nonparametric | interval/ratio, nominal/ordinal |
| significance def & formula | p<.05, how much faith we have in how right we are |
| liberal stat tests increase | Type I error, p is .2 means 20% chance of being wrong, anything over p<.05 |
| inferential stats | generalize from sample to pop., test hypotheses |
| regression analysis | allows us to predict (ex, super bowl predictions) |
| Pearson's product | 2 interval/ratio |
| point-biserial | 1 nominal, ratio/interval |
| phi | 2 nominal |
| Spearman’s rank order | ordinal correlations |
| Third variable problem and how to fix (partial correlations) | outside 3 variable, fix- partial correlations |
| mistakes we make in correlations | causality and directionality |
| Restrictive range of motion | cutoffs, ex, GRE cutoff cores |
| quantitative vs qualitative | change in quantity; categorical, what is your experience? |
| Absolute value of a correlation | magnitude (strength) use absolute value |
| Frequency distribution | tables |
| Z-scores are most affected by | SD |
| range is the... | simplest measure of variation |
| SD/average deviation is... | sophisticated |
| Disadvantage of using the range | limited, easily distorted by high or low scores |
Created by:
Finch35i