| A | B |
|---|
| ANALYTICS | using statistics to analyze the big data resulting from data mining |
| BAR CHART | a chart which can be used with most scales of measurement |
| CONFOUNDING VARIABLE | another variable which can influence the dependent variable and make it difficult to interpret the impact of the independent variable |
| CONTINGENCY TABLE | a cross tabulation for showing the relationship between two variables |
| CONTROL GROUP | in an experiment, this group is compared with the experimental group |
| PLACEBO | this is what is given to the control group in order to equalize expectations |
| CORRELATION | the relationship between two variables |
| POSITIVE CORRELATION | a direct relationship between two variables |
| NEGATIVE CORRELATION | an inverse relationship between two variables |
| LETTER E | this letter stands for scientific notation (a very large or very small number) |
| LETTER N | this letter stands for the number of participants in our sample |
| LETTER R | this letter stands for correlation coefficient, especially a Pearson product moment correlation coefficient |
| LETTER P | this letter stands for the probability of the null hypothesis |
| EXPERIMENT | this research is best for inferring causation because it manipulates an independent variable |
| INFERENTIAL | statistics which allow us to calculate or estimate the probability of the null hypothesis |
| LIKERT | this ordinal scale measures an attitude in terms of level of agreement |
| LINE GRAPH | this graph is useful in a time series |
| MEDIAN | this measure of central tendency is appropriate for a skewed distribution |
| NOMINAL SCALE | this scale measures a variable by categorizing responses |
| ORDINAL SCALE | this scale measures a variable by ranking responses |
| NONPARAMETRIC | these inferential statistics are appropriate when the data set is not normally distributed |
| NULL HYPOTHESIS | this explanation should be accepted if the results could be attributed to pure chance |
| PIE CHART | this circular chart is appropriate for nominal and ordinal scales when there are few categories or levels |
| SKEW | this term defines a non-symmetrical data distribution |
| QUASI-EXPERIMENT | this is a separate groups design in which the grouping is not based upon random assignment |
| RANDOM ASSIGNMENT | when each subject in the sample has an equal chance (compared to every other subject) of being assigned to the experimental group |
| RATIO SCALE | when a variable is measured by a score reflecting a number; the scale has a true zero point and proportionate scaling |
| RELIABILITY | consistency of measurement |
| REPEATED MEASURES DESIGN | a research design in which each subject receives more than one measure per variable, e.g., before/after, matched pairs |
| SAMPLE VS. NORMS | a research design in which the entire sample is compared to some external norms, e.g., census data |
| SCATTERPLOT | a chart appropriate for showing a correlation between bivariate data |
| SEPARATE GROUPS | a research design in which two (or more) parts of the sample are compared to each other |
| STATISTICAL SIGNIFICANCE | when inferential statistics determine that the results were probably not produced by pure chance |
| SPURIOUS CORRELATION | a correlation between two dependent variables due to them both being the results of the same independent variable |
| VALIDITY | when a test measures what it is supposed to measure |
