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;
53 Cards in this Set
- Front
- Back
Bar Graph |
commonly grapsh frequency distributions separate bar for each pieces of information can compare group means of percentages |
|
Pie Chart |
used when comparing group percentages /nominal information |
|
FREQUENCY pOLYGON |
graphs data using lines to represent each group best fro interval or ratio scales |
|
Histogram |
display frequency distribtuion where variable is measured in continous values used for quantitative variable |
|
Central Tendedency |
tells what sample is like as a whole meausred by mean, median, mode |
|
median |
divides group in half. 50 above, 50 below average line |
|
Mode |
most frequent score |
|
Variability |
amount of spread in distribution of scores measured by standard deviation |
|
Standard Deviation |
how far scores lie from the mean |
|
Variance |
standard deviation squared |
|
Correlation Co-Efficient |
how strongly variables are related to one antoher |
|
Pearson product Correlation Coefficient |
strength and direction of a relationship between variables |
|
restriction of range |
occurs when individuals being sampled are very similar on the variable you are studying. should be avoided |
|
Effect Size |
strenght of association between variables indicated by correlation coefficient or cohens d |
|
Cohen's D |
used when there are two or more treatment conditions shows effect size in units of standard deviation eg. d of .2 tells you means are separated by .2 standard deviations no maximum value. minimum is zero |
|
Regression Equations |
predict's person's score on one variable when score on another is already known |
|
Multiple Correaltion |
comnines number of predictor variables to increase accuracy of prediction of outcome =R R^2 represents amount of variability |
|
partial correlation |
statistically controls third variables in a non experiment |
|
Threats to Internal Validity |
-history effects -maturation -instrument decay -regression towards the mean -mortality - |
|
Multiple Baseline Across Subjects |
-different treatments are applied at different times to multiple subjects |
|
Interrupted Time Series VS. Control Series Design |
interrupted- measure, apply variable, continue to measure control series involves same procedure but there is a comparison group |
|
Pv |
participant variable eg. weight, gender |
|
Levels |
conditions in a factorial design |
|
Mixed Factorial Design |
Includes both repeated measure (within subjects) and independent groups (within groups) |
|
Moderator Variable |
Creates an interaction between variables in a factorial design |
|
Simple Main Effect |
differences between levels of IV as though there were different experiments at each level of IV |
|
Multiple Level Design |
multiples levels of IV but still only one variable |
|
Benefits of Factorial Design |
consider effects of more than on IV on DV see how IV's affect one another reeals interactions |
|
2X 3 X 2 |
2 IVs 2 with 2 levels, one with 3 |
|
Main Effect |
To find main effect, ignore the other variable |
|
What indicates interaction in factorial design? |
Lines that are not parrallel |
|
IV X PV Design |
how different types of people respond to same manipulated variable |
|
Independent Groups Factorial design |
2 X 2 Design Different groups of people in each condition |
|
Repeated Measures Design |
Same group of people used in each condition |
|
Combined Repeated Measures and Independent Group |
One variable uses same group for each condition other uses different groups
|
|
Straight Forward vs. Staged Manipulations |
staged-manipulation of environment or psychological state, stimulates real world situation straightforward-variables presented as text |
|
Asch Conformity Study |
confederate gives wrong answer when asked which line is longer |
|
Why use strong manipulation of IV? What are some downsides? |
maxmizes chance of finding a difference statistical significance: more likely that the difference is real as opposed to random error variation COns: may be to strong of an IV, may not reflect real world |
|
What are the three types of DVs? |
self report behavioural-direct observation physiological- GSR, EMG, EEG, mri |
|
Inter-Rater Reliability |
Degree of accordance between raters observations |
|
Low Grade vs Elaborate Deception |
low grade, hide purpose of study elaborate, avoids contamination of study due to subjects expectancies |
|
Balanced Placebo |
decpetion 50 get alcohol, 50 placebo 25 are told alcohol, 25 told no alcohol double blind |
|
Independent/ Between Groups Design |
Different participants are assigned to different groups |
|
Validity |
Are we really measuring what we claim to measure are there any confounding variables |
|
Key Features of Posttest Only design |
group assignment is assumed to be random as long as sample size is large enough control group reduces other confounds |
|
Multilevel, Randomized between subjects design |
adds more groups |
|
Solomon's Four Group Design |
reveals whether or not the pretest is acting as a confounding variable all undergo post test
grp a pre treatment post grap b pre no treatment post grp c treatment post grp d no treatment post |
|
Counterbalancing |
All possible orders of the presentation of variables/stimulus are applied. this reduces practice effects. to determine possible amount of orders factorial numbers 3!= 3 x 2 x 1 |
|
Latin Square |
partial counterbalancing |
|
Matched Pairs |
match people firs ton participant characteristic then randomly assign random person from each pair to a condition reduces error variance |
|
Structural Equation Model |
specifies a se of relationships among variables wen using nonexperimental method compare data to expteced ptern set outby model model is base don theory of how variabls will be related
|
|
Difference between Inferential and Descriptive Sas |
Inferential: extrapolates sample population data to a larger population Descriptive: reveals nature of the data, quantitatively describes data, often uses graphs or charts |
|
Power |
Probability that the nul will be rejected correctly. Higher power means less likely to obtian false negative (type II error) |