Eppp Statistics

Front 1

What are the four scales of measurement?

Back 1

Nominal
Ordinal
Interval
Ratio

Front 2

What are the measures of central tendency?

Back 2

Mean
Median
Mode

Front 3

What are the measures of variability (dispersion)?

Back 3

Variability
Range
Variance
Standard Deviation

Front 4

What is the relationship between variance and standard deviation?

Back 4

Variance = Standard Deviation squared

Front 5

What are methods of sampling?

Back 5

Interval recording
Event sampling
Situational sampling
Sequential analysis

Front 6

What is interval recording and when should you use it?

Back 6

Interval recording involves observing behavior during an interval of time.

It is best for studying complex interactions with no clear beginning or end (e.g., laughing, talking, playing)

Front 7

What is event sampling and when should you use it?

Back 7

Observing behavior each time it occurs.

Good for studying behaviors that occur infrequently, long duration, leave permanent record.

Front 8

What is situational sampling and when should you use it?

Back 8

Observe a behavior in a number of settings.

Increases generalizability.

Front 9

What is sequential sampling and when should you use it?

Back 9

Coding sequences of behavior rather than isolated behaviors.

Used to study complex behaviors.

Front 10

What are three sampling techniques?

Back 10

Simple random sampling - every member of population has an equal chance of being selected.
Stratified random sampling - dividing into strata and then randomly assigning
Cluster sampling - selecting clusters rather than individuals

Front 11

How can you minimize systematic error?

Back 11

1) Random assignment
2) Hold confounding variable constant
3) Match subjects on confounding variable
4) Blocking (build variable into study)
5) Statistically control confounding variable

Front 12

What are threats to internal validity?

Back 12

1) Maturation
2) History
3) Testing
4) Instrumentation
5) Statistical regression
6) Selection
7) Attrition
8) Interaction with selection

Front 13

What are threats to external validity?

Back 13

1) Interaction between testing & treatment
2) Interaction between selection & treatment
3) Reactivity (demand characteristics and experimenter expectancy)
4) Multiple treatment effects (order effects, carry over)

Front 14

What are the type of group designs for experimental research?

Back 14

1) Between groups design (factorial for 2+ IV)
2) Within subjects design (problem with autocorrelation)
3) Mixed design (both within and between)

Front 15

What are the three kinds of single subject designs?

Back 15

1) AB Design
2) Reversal Design (ABA)
3) Multiple baseline design

Front 16

What are the three predictions of the Central Limit Theorem?

Back 16

1) As sample increases, distribution approaches normal.
2) Mean of sampling distribution is equal to population mean.
3) SD of sampling distribution is equal to the population SD divided by the square root of sample size

Front 17

How can you increase statistical power?

Back 17

1) Increase alpha
2) Increase sample size
3) Increase effect size
4) Minimize error
5) Use one-tail test
6) Use parametric test

Front 18

What are the tests for nominal data?

Back 18

1) Single sample Chi-Square (1 variable)
2) Multiple-sample Chi-Square (2+ variables)

Front 19

What are the tests for ordinal data?

Back 19

1) Mann-Whitney U Test (1 IV, 2 indep groups; 1 DV rank-ordered)
2) Wilcoxon Matched-Pairs Signed Ranks Test (1 IV, 2 correlated groups; 1 DV rank-ordered)
3) Kruskal-Wallis Test (1 IV, 2+ indep groups; 1DV, rank ordered)

Front 20

What are the tests for interval or ratio data?

Back 20

1) Student t-test (1 IV; 1 DV) - single sample, independent sample, and correlated sample
2) ANOVA (1+ IV, 2+ indep groups; 1 DV) - one way, factorial, randomized block factorial, ANCOVA, repeated measures, mixed (split-plot), MANOVA
3) Trend analysis

Front 21

What is effect size?

Back 21

Effect size is a way to evaluate the clinical/practical significance of statistically significant results. It is also used in meta-analysis.
1) Cohen's d - diff between two groups
2) r square and eta square - amount of variance in outcome variable accounted for by variance in treatment

Front 22

What are the different correlation coefficients?

Back 22

1) Pearson f (both i/r)
2) Spearman rank order (rho) (both ranked)
3) Phi (both true dichotomy)
4) Tetrachoric (both artificial dichotomy)
5) Contingency (both nominal)
6) Point biserial (true dichotomy, i/r)
7) Biserial (artificial dichotomy, i/r)
8) Eta (nonlinear, both i/r)

Front 23

What are the three major assumptions for correlation coefficients?

Back 23

1) Linearity
2) Unrestricted range
3) Homoscedasticity (same range for x as y)

Front 24

What are the multivariate techniques?

Back 24

1) Multiple regression (2+ continuous IV; 1 continuous DV)
2) Canonical correlation (2+ continuous predictors, 2+ continuous criterion)
3) Discriminant function analysis (2+ continuous predictor, 1 nominal criterion)
4) Logistic Regression (nonlinear version of #3)