Term
| Independent Variable (IV) |
|
Definition
| Known as the predictor or causal variable |
|
|
Term
|
Definition
| Known as the criterion or outcome variable |
|
|
Term
|
Definition
This type of analysis is where there is one or more independent variable/s (IV) and only one dependent variable |
|
|
Term
|
Definition
| This type of analysis is where there are two or more indpendent variables (IVs) and only one dependent variables |
|
|
Term
|
Definition
| This type of analysis is where there are two variables and neither are listed as the IV or DV |
|
|
Term
|
Definition
| This type of analysis is where there are one or more IVs and two or more DVs |
|
|
Term
|
Definition
| This measure of central tendency is best for variables measured on a nominal level. |
|
|
Term
|
Definition
| This measure of central tendency is best when extreme scores are not considered and is appropriate for ordinal variables |
|
|
Term
|
Definition
| This measure of central tendency takes into account all scores in the distribution and most appropriate when variables are interval or ratio. |
|
|
Term
|
Definition
| This measure of relative position indicates the percentage of scores that fall at or below a given score |
|
|
Term
|
Definition
| This measure of relative position is derived from the manipulation of a raw score that expresses how far away from the mean a give score is located, usually reported in terms of standard deviation |
|
|
Term
|
Definition
| This standard score indicates the distance away from the mean score in terms of standard deviationunits and serves as a quick indicator of whether the score is located above or below the mean |
|
|
Term
|
Definition
| This measure of relationship is used if data for one or both variables are ranked (ordinal) instead of quantitative scores. |
|
|
Term
|
Definition
| If both variables are interval or ratio, this measure of relationship is appropriate b/c it takes into account the value of every score in both distributions. |
|
|
Term
|
Definition
| Any sample will, in all likelihood, not perfectly represent the population. This expected, chance variation among sample means is known as: |
|
|
Term
| Distibution of Sample Means |
|
Definition
| . If enough samples are selected and means are calculated for each sample, all samples will not have the same mean, but those means will be normally distributed around the population mean – this is called the: |
|
|
Term
|
Definition
| . The standard deviation of the sample means is usually referred to as the ________. This tells us by how much we would expect our sample means to differ if we used other samples from the same population – and how well our sample represents the population from which it was selected – the smaller the better! With a smaller ____________ we can have more confidence in our inferences we make about the population based on the sample data. |
|
|
Term
|
Definition
| States that there is no true difference to be found in the population |
|
|
Term
|
Definition
| If the researcher rejects the null hypothesis but is wrong |
|
|
Term
|
Definition
| if a null hypothesis is false but the researcher concludes that it is true |
|
|
Term
|
Definition
Also known as the Alpha level α- a pre-established (a priori) probability value of being incorrect.
(The social sciences most commonly use α = .05 → the percentage we risk being wrong.) |
|
|
Term
|
Definition
| The probability of rejecting the null hypothesis when it is in fact false (a correct decision). Like alpha, this is established arbitrarily, and should be set at a high level. |
|
|
Term
|
Definition
This is the size of the treatment effect the researcher wishes to detect with respect to a given level of power. A more powerful statistical test will be able to detect a smaller ___________.
.2 is considered small, .5 is considered medium, and .8 is considered large. |
|
|
Term
|
Definition
This takes into account things such as how many participants were involved in the research and the strength of the research (how significant they were), termed "d"
A higher ______ indicates a stronger result – means more ppl were affected |
|
|
Term
|
Definition
| The most commonly used bivariate correlation technique – it measures the association between 2 quantitative variables without distinction between the IV and the DV. |
|
|
Term
| Bivariate (Linear) Regression |
|
Definition
This analysis utilizes the relationship between the IV and DV to predict the score of the DV from the IV.
(1 IV and 1 DV and we’re trying to determine if the IV will predict the DV) |
|
|
Term
|
Definition
| This analysis identifies the best combination of predictors (IVs) of the DV – it’s used when there are several independent quantitative variables and 1 dependent quantitative variable. |
|
|
Term
|
Definition
| This analysis selects IVs, one at a time, by their ability to account for the most variance in the DV. As a variable is selected and entered into the group of predictors, the relationship between the group of predictors and the DV is reassessed. |
|
|
Term
|
Definition
| 2+ IVs (quantitative) → measure of relationship/prediction 1 DV (quantitative) Which analysis is this? |
|
|
Term
|
Definition
A job promotion is the DV and our IVs are age, experience, and gender. We’re trying to find the best combination of IVs to predict the DV (promotion).
Name the analysis you would use. |
|
|
Term
|
Definition
The most basic statistical test that measures group differences |
|
|
Term
|
Definition
This analysis analyzes significant differences between 2 group means. 1 IV (2 categories or levels ) i.e. yes or no → group differences 1 DV (quantitative) |
|
|
Term
|
Definition
| This analysis tests the significance of group differences between 2 or more group means, as it analyzes variation between and within each group. This analysis is appropriate when the IV is defined as having 2+ categories (levels) and the DV is quantitative. |
|
|
Term
|
Definition
1 IV (2+ categories/levels) → group difference 1 DV (quantitative) Name the analysis. |
|
|
Term
|
Definition
The IV is selection methods and levels are banding, top down, and GMA test. The DV is job performance. We want to know is there are any diff amongst the three levels.
Name the analysis you would use. |
|
|
Term
|
Definition
2 or more groups are being compared on the mean of some DV, but this analysis additionally controls for a variable (covariate) that may influence the DV. Name the analysis. |
|
|
Term
|
Definition
1 IV (2+ categories/levels) → group differences 1 DV (quantitative) 1+ covariate
Name the analysis. |
|
|
Term
|
Definition
We're looking at education and income – Does edu affect income? What about job experience? It might have an affect on what we’re looking at.
After accounting for experience there may or may not be a relationship, however at least we’ve accounted for any confounding variables or anything that could be influential.
What analysis would you use? |
|
|
Term
|
Definition
| In ANCOVA and MANCOVA, we don't want the covariate to affect our relationship because that would then result in a ________________. |
|
|
Term
|
Definition
| This analysis is utilized to simultaneously study 2+ related DVs while controlling for the correlations among the DVs. |
|
|
Term
|
Definition
1 IV (2+ categories/levels) → group differences 2+ DVs (quantitative) |
|
|
Term
|
Definition
| This analysis investigates group differences among several DVs while also controlling for covariate(s) that may influence the DVs. |
|
|
Term
|
Definition
1 IV (2+ categories nominal or coded) → group differences 2+ DVs (quantitative) 1+ covariate Name this analysis. |
|
|
Term
|
Definition
This analysis is seen as the reverse of MANOVA in that it seeks to identify which combination of quantitative IVs best predicts group membership as defined by a single DV that has 2+ categories |
|
|
Term
|
Definition
2+ IVs (qualitative) 1 DV (2+ categories) → group prediction Name this analysis. |
|
|
Term
|
Definition
| This analysis allows the researcher to explore underlying structures of an instrument or data set and is often used to develop and test theory. |
|
|
Term
|
Definition
| This analysis combines several related IVs into fewer, more basic underlying groups. IVs that share common variance are grouped together. Once these groups are created, they are often adjusted (rotated) so that these groups are not highly related to one another and more accurately represent the combined IVs. |
|
|
Term
|
Definition
| ___________ provides a direct indication of the amount of inconsistency or error to be expected in an individual’s score. |
|
|
Term
|
Definition
| This statistical test allows us to test for deviations of observed frequencies from expected frequencies. |
|
|
Term
|
Definition
| A measure of the degree to which the scores in a distribution are clustered together or spread apart. |
|
|
Term
|
Definition
| The number of scores in a sample that are free to vary with no restriction. |
|
|
Term
|
Definition
| A standardized score with a sign that indicates direction from the mean (+/-) and a numerical value equal to the distance from the mean measured in standard deviations. |
|
|
Term
|
Definition
| The standard deviation of the distribution of sample means. The standard distance between an sample mean and the population mean |
|
|
Term
|
Definition
_____ is also known as the probability of a Type II error. |
|
|
Term
|
Definition
| This statistic is used to summarize sample data in situations wherre the population standard deviation is not known. This statistic uses an estimate of the standard error. |
|
|
Term
|
Definition
| An interval estimate that is described in terms of the level (percentage) of confidence in the accuracy of the estimation. |
|
|
Term
|
Definition
| The test statistic used for ANOVA- compares the difference (variance) between treatments with the differences (variance) that are expected by chance. |
|
|
Term
|
Definition
| A test that is conducted after an ANOVA with more than two treatment condidtions where the null hypothesis was rejected. The prupose of this test is to determine eactly which treatment conditions are significantly different. |
|
|
Term
|
Definition
| Unexplained, unsystematic differences that are not caused by any known factor |
|
|
Term
|
Definition
| The overall mean differences between the levels of one factor. When the data are organized in a matrix, the __________ are the mean differences among the rows (or among the columns). |
|
|
Term
|
Definition
| Mean differences that cannot be explained by the main effects of the two factors. An _________ exists when the effects of one factor depend on the levels of the second factor. |
|
|
Term
|
Definition
| Hypothetical, ideal frequencies that are predicted from the null hypothesis |
|
|
Term
|
Definition
| The actual frequencies that are found in the sample data |
|
|
Term
|
Definition
When samples have approximately equal variances, they are said to be ________. This is an assumption of several analyses. |
|
|
Term
|
Definition
| When choosing which t-test to use, this t-test should be used when there is no correlation between the two groups |
|
|
Term
|
Definition
| When choosing which t-test to use, you would use this t-test when two sets of data (pair of scores) from matched subjects or from the same subject (repeated measures) are correlated. |
|
|
Term
|
Definition
| This statistic calculates mean difference/standard error. |
|
|
Term
|
Definition
| By maximizing mean difference and minimizing standard error, you are increasing _______. |
|
|
Term
|
Definition
| This type of kurtosis occurs when the value is over +2, indicated a non-normal distribution. |
|
|
Term
|
Definition
| This type of kurtosis occurs when the value is below -2, indicates a non-normal distribution. |
|
|
Term
|
Definition
| Considers whether a statisticall significant research finding has applied value and use, can be important when deciding whether or not to act upon a research finding. |
|
|
Term
|
Definition
| Traditionally _________ should explain 20% of the total variance, can be tested using several formulas, the most common is eta squared |
|
|
Term
|
Definition
| Based on a predetermined level of probability corresponding to random findings established by the researcher prior to conducting a chosen analysis, varies as a function of the discipline (Business= .10, Psychology = .05, Medical = .01) The probability corresponds to the possibility that the finding occured randomly. Thus, when the _______ is .05, the likelihood of random findings is 1 out of 20 and the likelihood of non-random findings is 19 out of 20. |
|
|
Term
| Effect Size, Alpha, Sample Size |
|
Definition
| Power is determined by these three factors: |
|
|
Term
|
Definition
The degree to which variables can be predicted or accounted for by other variables in the analysis- As this increases, the ability to define a single variables' effect is diminished. |
|
|
Term
|
Definition
| A distribution is said to be ________, when there is a skewed distribution, resulting in unequal dispersion or from sample size discrepencies |
|
|
