Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
44 Cards in this Set
- Front
- Back
- 3rd side (hint)
If a confound is present and the DV varies in different conditions |
The experimenter is Not able to determine whether the changes were caused by the IV (s) or by the confound |
|
|
Confounds influnce |
The measures of central tendency for different conditions |
|
|
Independent-sample design |
Random assignment, keep conditions other than IV the same, Distribute potential confounders across conditions |
|
|
Dependent sample design |
Counterbalancing Balances order or practice effects |
|
|
Advantages of more conditions |
Efficiently compare multiple conditions, Less chance of type1Error ( compared to two group exp ..) |
|
|
Comparing T-test and anova |
Independent samples t -randomly selected samples, dv normally distributed, dv measured using ratio or interval scale, homogeneity of variance, independent groups... Independent Anova- randomly selected. Dv normally distributed, dv measured using ratio or interval scale, homogeneity of variance, independent groups |
Both have the same comparison |
|
Comparing stat hypotheses independent t-test and anova (population differences) |
T test- = or not = Anova- not all (u's) are equal or at least population mean is different from another |
|
|
Advantage of one way anova test over multiple t test |
Reduce type 1 error |
|
|
Type 1 error |
When we reject the null hypothesis when its true Can only occur if results of inferential test show significant difference Occurs when we say a manipulation had an effect when it did not The experimenter directly control the type 1error risk by selecting a alpha level |
|
|
The lower the alpha level the .... |
Lower the risk of type 1 error |
|
|
The more test you do the. ... |
The greater the risk of type 1 error |
|
|
Anova reduces the probability of a type ...... ____ error |
1 ( type 1 error) |
|
|
Experiment-wise error rate |
Is the type 1error rate for entire set if inferential tests Probability that at least one of the t-test will have type 1error Anova allows making many comparisons while keeping the experiment-wise error @.05 |
|
|
Bonferroni adjustment |
Setting lower alpha level for many tests to reduce type 1 error Not deal- increases type 2 error ( reducing power) |
|
|
Independent samples are also called |
Randomized anova |
|
|
Independent samples are also called |
Randomized anova |
|
|
What is anova |
Inferential stats to determine if there are differences between more than 2 groups
Resulting in F statistic
|
|
|
What do we calculate in anovav |
F statistic (obt) |
|
|
What is f statistic |
Ratio compares variance between sample means to the within |
|
|
Mean square (MS) in anovabis used for |
Variance |
|
|
Formula for Fobt |
Ms between /ms within |
|
|
Formula for Fobt |
Ms between /ms within |
|
|
Between sample |
Reflecting that a treatment was used |
|
|
Within sample |
Treated similarly or the groups are similar |
|
|
Both f and t statistic are |
Ratios if obtained difference (numerator) to the expected difference by chance if null hypothesis was true (denominator) |
|
|
T= |
Actual obtained difference between 2 sample means / estimated standard error ( difference expected with no treatment effect) |
|
|
How do we partition variance the total variability |
Variance between treatments And Variance within treatments |
|
|
Between groups variability |
Variability caused by independent variable |
|
|
Within groups variability /(error variability) |
Variability due to factors such as individual differences Errors in measurement In differences nit due to independent v |
|
|
The lower the error variability the greater the... |
F statistic |
|
|
The lower the error variability the greater the... |
F statistic |
|
|
F ratio if no treatment effect ( f ratio would be close to 1) |
F= 0+ differences due to chance/ differences due to chance |
|
|
F distribution is ..._____ skewed |
Positively |
|
|
F distribution is ..._____ skewed |
Positively |
|
|
Sig f ratio indicates?! |
Unlikely to occurred by chance |
|
|
Use what test to determine which means differ |
Post hoc ( follow up tests) |
|
|
Use what test to determine which means differ |
Post hoc ( follow up tests) |
|
|
Tukey is what kind of test |
Pos hoc |
|
|
What is tukey used for |
Pairwise comparisons. . formula calls for dividing the differences between pairs by means of standard error |
|
|
Type 1error |
Rejecting null when its true |
|
|
Type 2 error |
Fail to reject the null when its false |
|
|
Type 1 error (false positive error) |
Null hypothesis is true , but was rejected ( proven untrue ) after testing |
|
|
Type 2 error ( false negative) |
Null hypothesis is false but (after testing) was proven true |
|
|
Partial counter balancing |
Randomized counter balancing Latin square |
|