Question | Answer |
mean | average-add up all the numbers; divide by number of numbers. Interval/ratio |
range | difference between largest and smallest numbers |
median | middle number-place in numerical order first. Ordinal |
standard deviation | score root of the variance of a set of numbers. Used with interval and ratio data. |
mode | number that occurs most often. Nominal |
ANOVA | analysis of variance. A simple one way analysis of variance (ANOVA) is a parametric test used to determine whether scores from 2 or more groups are significantly different at a selected probality level. |
statistical error | difference between estimated and true value (random) |
systematic error | same as above (non-random) |
type I error | false positive - rejecting null hypothesis when true |
type II error | false negative - the error of failing to reject a null hypothesis when it is not true |
variance | a measure of the amount of variance within the values of that variable |
t test | Used to determine whether two groups of scores are significantly different from one another. It compares teh observed difference between means with the difference expected by chance. |
wilcoxon signed rank | designed to test a hypothesis about the location of a population distribution |
quartile divisions | one of the four divisions of observations which have been grouped into 4 equal sized sets based on their statistical rank |
descriptive statistics | describing what is or what the data shows |
inferential statistics | trying to reach conclusions that extend beyond the immediate data alone |
problem statement | indicates particular focus of study |
review of the literature | comprehensive, flows, relevant, primary sources, cited accurately |
hypotheses | specific, testable, expected relation |
particpants | described, adequate size, no sampling bias |
instruments | comply with IRB standards, appropriate, rational |
research design and procedures | appropriate, applied correctly, logically relates |
results | descriptive statistics presents, assumptions met |
discussion | each result discussed, generalizations consistent with results |
abstract and summary | problem stated, design identified, procedures described |
correlational | a measure of the association of 2 or more dependent variables |
Causal comparative | attempts to identify causative relationship between independent/dependent variable |
Experimental | =maintains control over all factors |
Correlational coefficient== | a linear relationship between 2 variables |
Pearson r | tells the magnitude of the assoc. between variables on a ratio scale. Used to calculate relationship for interval or ratio data. |
Prediction study | determines which variables are most highly correlated w/criterion variable |
content validity | addresses whether the test measures the intended content area |
criterion related validity | determined by relating performance on a test to performance on an alternative test or other measure |
construct validity | assesses what the test is actually measuring |
consequential validity | refers to the extent to which an instrument creates harmful effects for the user |
stability test retest | empirical item - selection-same group/same test measure the stability of scores over time |
equivalence:equivalant forms | equality-same group of people/different forms of test |
internal consistency | a measure based on the correlations between items on the same test |
split half | randomly divide all items into two sets |
scorer/rater reliabilty | determines consistency in which rater asigns scores |
If a test is reliable, does it also mean that it is valid? | Not necessarily |
If a test is valid, is it also reliable? | Yes |
Spearman rho | Used to calculate relationship with ordinal data. |
dependent variables | those believed to depend on or to be caused by another variable |
independent variables | the hypothesized cause of the dependent variable |
Null Hypothesis | there is little existing research or theoretical support for a hypotheisi |
Chi square | a nonparametric test used to test differences between groups when the data are frequency counts or the percentages or proportions converted into frequencies |
ethnography | the form of qualitative research that focuses on describing the culture of a group of people |
case study research | form of qualitative research that is focused on providing a deatiled account of one or more cases. |
Grounded theory | a qualitative approach to generating and developing a theory from data that the researcher collects. |
Nominal Measurement | the numerical values just name the attribute uniquely. Jersey numbers in basketball, for example. Qualitative |
Ordinal Measurement | Attributes can be rank ordered. Distances between attributes have no meaning. Years of college on a questionaire.Quantitative |
Interval Measurement | Differences between the attribues Does have meaning. Measuring temperature.Interval between values is interpretable. Quantitative |
Ratio Measurement | ABSOLUTE ZERO that is meaningful. Number of clients in 6 months. Quantitative |
raw score | number of items correct |
norm-refernced scoring | student's performance compared with performance of others |
Criterion refernced scoring | student performance compared to preset standard |
self-refernced scoring | how individual's scores change over time |
performance assessment | emphasize student process and require a project |
cognitive tests | measure intellectual processes (thinking, memorizing) |
diagnostic tests | provide scores to facilitate identification of strengths and weaknesses (tests for SDC classes) |
aptitude tests | measure prediction or potential versus what has been learned (wescler) |
affective test | measure attitude, emotion, (Likert) |
threats to validity | unclear directions, confusing or unclear items, vocabulary or reading too difficult, subjective scoring, cheating, errors in administration |
sample | a set of elements taken from a larger population |
statistic | a numerical characteristic of a sample |
parameter | a numerical characteristic of population |
random sampling | produces representative samples |
non-random sampling | does not produce representative samples |
EPSEM | equal probability sampling method |
systematic sampling, | SELECT EVERY NTH NUMBER IN YOUR SAMPLE, EPSEM |
stratified random sampling | divide into males/females, take a random sample from each |
proportional stratified samplng | EPSEM, subsamples (males/females) must be proportional to size in population |
disproportional stratified sampling | not prportional to size in population |
one stage cluster sampling | select random sample of clusters (classrooms) then include all individual units in the selected clusters |
two stage cluster sampling | same as above except take subsets (like 10 kids from each classroom |
convenience sampling | people most available nonrandom |
quaota sampling | setting quotas (25 Asians, 25 African Americans, etc) nonrandom |
purposive sampling | researcher specifies characteristics (7th graders with ADD)nonrandom |
snowball sampling | start with one, ask them to find one, ask them to find one, etc. nonrandom |
maximum variation sampling | wide range of cases qualitative |
homogeneous sample selection | small and homogeneous case or set of cases qualitative |
extreme case sampling | cases that represent extremes on some dimensions |
typical-case sampling | typical or average cases |
critical case sampling | cases known to be very important |
negative case sampling | cases that diconform your generalizations |
opportunistic sampling | select useful cases as the opportunity arises |
mixed purposeful sampling | mix sampling strategies tailored to your specific needs |