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;
68 Cards in this Set
- Front
- Back
Public Law 93-380
|
Buckley Amend- Confidentiality of school and college info. May not even release to parents unless under 18
|
|
Research
|
Collecting and analyzing info about a part subject
Deductive-prove or disprove a theory Inductive- dev a theory to show patterns in data |
|
Quantitative vs Qualitative research
|
Quantitative- systematic using scientific methods. Hypothesis etc
Qualitative- in depth |
|
Internal validity
|
The extent to which the result of the exp can be attributed to the variable
|
|
External validity
|
The accuracy to which the exp can be generalized to a larger pop
|
|
Hawthorne effect
|
Subjects knowing they are involved in a study and the atten they get
|
|
Rosenthal or Pygmalion effect
|
Subjects change their behavior bc of att, expec, or beh of the researcher
|
|
Demand characteristics
|
Info received by the subjects Inc rumors they heard about the study they're in
|
|
Levels of measurement
|
NOIR
Nominal, ordinal, interval, ratio |
|
Nominal scale
|
Categorical Variables gender, race etc
|
|
Ordinal measurement
|
Variables in order or ranking
|
|
Interval measurement
|
Variables with similar or equal distances bwt ranks generations, crime rates
|
|
Ratio measurement
|
Equal interval, 0 reference. Little used in the social sciences since attitudes not measured at 0
|
|
Sampling
|
Selecting a part of a population to make a generalization
|
|
Types of sampling
|
Random-chosen by chance
Stratified- divides into subgroups acc to criteria. Good to generalize Proportional- the selection of the # of subjects from each subgroup corresp to pop Cluster-divides into subgroups then selects randomly from them Purposeful- selection for in-depth study |
|
Meta-analysis
|
Answering a research quest through the comparison of results from multiple studies
|
|
Sample size
|
# of samples in a study
|
|
Table of random numbers
|
List of random # (usually comp generated) that can be assigned to potential study samples and used to randomly select who will part
|
|
Likert scale
|
Rating scale on which part agree or disagree w statements that measure attitudes or opinions
|
|
Scatterplot
|
Graphic using horizontal and vertical lines to illustrate the relationship btw 2 variables
|
|
Type 1 error
|
Alpha error- null hypothesis is rejected as false when actually true
Most damaging |
|
Type 2 error
|
Beta error- null hypothesis is accepted when it's actually false
No relationship when there is one *often result of sample sz too small |
|
Hypothesis
|
Prediction or statement that will be shown by a study
|
|
Null hypothesis
|
Assumes No relationship btw variables
|
|
Directional hypothesis
|
Predicts how the ind variable will affect the dep variable
|
|
Non-directed hypothesis
|
Predicts an effect but does not state how the dep variable will be affected
|
|
Significance level
|
Probability of making a type 1 error in a hypothesis test
Low- .05 or 5% usually being used |
|
T-Test
|
Compares the mean of 2 ind data sets to determine if there's a sign statistical diff btw them
-est table of t values - existence of relation btw data sets before standard deviation value determined - good for small grps |
|
Chi- square
|
Used to determine if there are significant diff in the distribution of 2 data sets
|
|
Bivariate analysis or crossbreak
|
Graphically illustrating the relationship or non-relationship of 2 variables by use of an X/Y graph
Ind var- vert axis Dep- horiz axis |
|
ANCOVA
|
Analysis of Covariance- dep variables are controlled ex non random sampling, stat adjust var that affect the dep variable
|
|
One-way analysis of variance
|
Test for diff when the study Inv 3 or more ind groups or levels
|
|
ANOVA
|
Study of 2 or more variables
2x2 most com |
|
MANCOVA
|
Several dep variables and at least 2 ind variables
|
|
Post hoc tests
|
Multiple comparison tests done after data sets in a study hv similar F values
|
|
Nonparametric tests
|
Validation tests when values are not distributed normally
2 samples are ind of each other Ordinal or nominal Median; spearman, Mann-Whitney |
|
Parametric tests
|
Normal distribution
Ratio or interval; mean; Pearson, t-test, anova Draw more conclusions |
|
Solomon four group design
|
Does the pre-test affect the subjects of a test by influencing
|
|
Multiple regression
|
Procedure where the researcher uses a correlation coefficient to learn about the rel btw multiple ind variables and a dep variable
"best predictor" |
|
Factor analysis
|
Rel among a group of variables for the simplest explanation usu the smallest # of factors
|
|
Biserial correlation coefficient
|
Relationship btw 1 variable w multiple values another that is dichotomous
|
|
Cross-sectional
|
Study of char of group over long period of time. Longitudinal study
|
|
Degrees of freedom
|
How many observ the researcher may make after the min is needed for the study
|
|
Double blind study
|
Neither researcher or subjects know at least 1 variable ex drug study
|
|
Homoscedasticity
|
Statistical variances are assumed equal
|
|
Heteroscedasticity
|
Unequal variance of data on either side of the regression line
|
|
Semantic differential
|
Method for measuring subjects reactions to words or concepts
|
|
Accountability
|
Effectiveness of treatment compared w just of the $
|
|
Formative evaluation
|
Effectiveness of treatment
|
|
Summative evaluation
|
How well the treatment meets it's goals
|
|
Types of bell curves
|
Leptocurtic- narrow, tall, symmetrical
Mesocurtic- normal Platykurtic- flat w a lot of frequency |
|
Positive skew
|
Tail to the right
|
|
Negative skew
|
Tail to the left
|
|
Range
|
Diff btw highest and lowest score
Sub highest from lowest |
|
Variance
|
The difference
Average variation of scores frm the mean. Calculate: 1. Calculate the mean 2. Subtract mean from each ind score 3. Square each of ans 4. Add all squared ans 5. Divide total frm the # of scores |
|
Standard deviation
|
Dividing a curve into equal score intervals
Square root of the variance |
|
Stanine
|
Method of scaling test scores on 9 pt standard scale
Mean of 5 SD of 2 |
|
Z-score
|
raw score minus mean divided by SD
Most basic Mean 0 SD 1 |
|
T-score
|
Most used
Mean 50 SD 10 |
|
Correlation Coefficient
|
Degree of a relationship btw variables.
Neg- one Inc the other dec Pos- both inc Most common: Pearson r; Spearman rho |
|
Inferential Statistics
|
Level of significance P confidence level: .05 or less then 5 or less times out of 100 results are not due to chance
|
|
Inferential statistics tests
|
Chi
T-test Anova |
|
Correlation
|
Relationship btw 2 or more variables
Linear or non- linear Pos or neg: -1.0-+1.0 0 no correlation Letter r Does not mean causation |
|
Pearson r vs spearman rho
|
Pearson r- interval or ratio
Spearman rho- ordinal (think row) |
|
Regression line
|
Best fitting line to minimize deviations
Used to predict y score |
|
Coefficient of determination
|
How much Y scores are due to regression line
|
|
Coefficient of non-determination
|
How much predicted y score is due to something else other than the regression line
|
|
SEM
|
Standard error of the measurement
|