| A variable that is measured to see whether the treatment or manipulation of the independent variable had an effect | What is a dependent variable? |
| A variable that is manipulated to examine its impact on a dependent variable | What is an independent variable? |
| A variable that is related to the dependent variable, the influence of which needs to be removed | What is a control variable |
| A variable that is related to the dependent variable or indepdendent variable that is not part of the experiment | What is an extraneous variable? |
| A variable that is related to the dependent variable or independent variable and has an impact on the dependent variable | What is a moderator variable? |
| Includes control group. Randomly selects participants from the population. Randomly assigns participants to groups. Randomly assigns treatments to groups (i.e, manipulate sthe independent variables). Has high degree of control over extraneous variables | What are the defining characteristics of a TRUE experiemental design? |
| The process where we take a meaningful and usually vague concept; and provide a way to measure it | What is operationalisation? |
| Provide reasonable answers to interesting questions: Ask the question, identify important factors, formulate hypothesis, collect relevant information, test hypothesis, work with hypothesis, reconsider the theory, ask new questions | What is the research process? |
| Assignment of labels; e.g., Gender, Preference (like or dislike), Voting record (for or against) | What are the qualities of nominal measurements? |
| Assignment of values along some underlying dimension: Rank in college, Order of finishing in a race | What are the qualities of ordinal measurements? |
| Equal distances between points: Number of words spelled correctly, Intelligence test scores, temperature | What are the qualities of interval measurements |
| Meaningful and nonarbitrary zero: Age, Weight, Time | What are the qualities of Ratio measurements? |
| Interval / Ratio measurements. Always has another value in between any two values (e.g., Time) | What is a continuous variable? |
| Nominal / Ordinal. Can have nothing in between two values (e.g., gender, race etc) | What is a discrete variable? |
| The reproducibility of the results/data obtained from a procedure or tool and the extent to which a measurement is consistent and free from error | What is Reliability? |
| Measure of stability: Administer the same test/measure at two different times to the same group of participants | Define test-retest |
| A measure of equivalence: Administer two different forms of the same test to the same group of participants | Define parallel-forms |
| A measure of agreement: Have two raters rate behaviours and then determine the amount of agreement between them | Definer Inter-rater |
| A measure of how consistently each item measures the same underlying construct: Correlate performance on each item with overall performance across participants (Cronbach's alpha / Kuder-Richardson) | Define Internal consistency |
| The accuracy, genuiness, and authenticity of your measure. | What is validity? |
| A measure of how well the items represent the entire universe of items: Ask an expert if the items assess what you want them to assess | Define Content validity |
| A measure of how well a test estimate a criterion: Select a criterion and correlate scores on the test with scores on the criterion in the present | Define concurrent Criterion Validity |
| A measure of how well a test predicts a criterion: Select a criterion and correlate scores on the test with scores on the criterion in the future | Define predictive criterion validity |
| A measure of how well a test assesses some underlying construct: Assess the underlying construct on which the test is based and correlate these scores with the test scores. | Define construct validity |
| By measures of central tendency, by measures of dispersion (variability) and by measures of association | How do we describe data? |
| The sum of a set of scores divided by the number of scores (average) | Define Mean |
| The score or the point in a distribution above which one-half of the scores lie (middle score) | Define Median |
| The score that occurs most frequently | Define Mode |
| A 'mean of the mean', it tells you how tightly all the various examples are clustered around the mean in a set of data | What is standard deviation? |
| Data is closely clustered around the mean | What does a low standard deviation look like? |
| Data is dispersed amongst a large range of numbers | What does a high standard deviation look like? |
| The mean, median and mode are all equal. +/- 1SD = 68%, +/- 2SD = 95% and +/- 3SD = 99% | What does a normal distribution look like? |
| Check any data entry errors BEFORE running analyses, gather info on descriptive statistics, identify patterns that may not otherwise be obvious, identify any missing data points and devise a strategy to deal with those. Identify and deal with sources of bias (e.g., outliers and violation of assumptions about data) | Why do we conduct exploratory data analysis? |
| Analyse -> Descriptive Statistics -> Frequencies -> can also click the options in the Statistics section - > Continue | Where do you go in SPSS to get descriptive statistics for categorical errors?? |
| Analyse -> descriptive statistics -> EXPLORE -> can also tick Outliers in the Statistics section -> Continue | Where do you go in SPSS to get descriptive statistics for Continuous variable errors? |
| Tenacity, Intuition, Authority, Rationalism and Empiricism | What is non-science? |
| Exclude cases listwise option will include cases in the analysis only if they have full data on all of the variables listed in your variables box for that case. The Exclude cases pairwise option however excludes the case (person) only if they are missing the data required for the specific analysis | Explain the difference between “exclude cases listwise” and “exclude cases pairwise” in dealing with missing data points. |
| Communicates how accurate the estimate is likely to be | What are confidence intervals? |
| Distance between smallest and largest value in the data set | Define Range |
| Dispersion of the values in the data set | Define Variance |
| Tells us what percentage of scores lie below a particular value. Represents a scores position relative to other scores. Divided into four quarters. 1st (lower) = 25th, 2nd (median) = 50th and 3rd (upper) = 75th | What is a percentile? |
| Q3 - Q1 | What is interquartile range? |
| Telll us if extreme scores are biasing our mean | What can a trimmed mean do? |
| Normality, Skewness and kurtosis | What information do you get from Histograms? |
| Look at continuous variable data across different groups (categorical variable data) | What information do you get from bar graphs? |
| Gives us information about measures of central tendency, outliers, and variability (including outliers) for continuous variables | What information do you get from box plots? |
| X-Axis | On which Axis does the IV sit? |
| Y-Axis | On which axis does the DV sit? |
| Scores bunched at low values with the tail pointing to high values | What is positive skew? |
| Scores bunched at high values with the tail pointing to low values | What is negative skew? |
| the distribution has heavier tails than the normal distribution, looks more peaked | What is postivie kurtosis? |
| The distribution has lighter tails than the normal distribution, look flatter | What is negative kurtosis? |
| Graphs -> Graph builder - > Select bar graphs in gallery tab -> Place categorical variable on the X-axis and continuous variable on the y-axis -> Tick display error bars -> Standard Error -> ok | How to generate bar graphs? |
| Provides visual information on outliers. Indicates the central tendency and variability in scores. Can plot for each group and compare | What does a box plot tell us? |
| An interval constructed such that the true population mean (M) will fall within this interval in 95% of samples | What is a confidence interval? |
| An option that will include cases in the analysis only if they have full data on all of the variables listed in your variables box for that case | Define exclude case listwise |
| Excludes the case (person) only if they are missing the data required for the specific analysis | Define exclude case pairwise |
| If you compare the original mean to it, you can see whether extreme scores are having a strong influence on the mean | Why is the 5% trimmed mean a useful statistic? |
| * symbol | What does an extreme score on a boxplot look like? |
| o symbol | What does an outlier score on a boxplot look like? |
| An outlier is a score very different from the rest of the data | What is an outlier? |
| Bias our paramter estimates, standard error and confidence intervals, and test statistics and p-values | What are three different ways we can bias our analysis? |
| Circle is an outlier and number is the row in the data editor | What is the little circle and number that's plotted ina box-plot? |
| Assumes underlying statistical distributions in the data. Therefore, several conditions of validity must be met so result of the test is reliable. E.g., Students t-test for two independent samples is reliable only if each sample follows a normal distribution and if sample variances are homogenous | What is a parametric test? |
| Does not rely on any distribution. They can be applied even if parametric conditions of validity are not met. | What is a non-parametric test? |
| Smooth and symmetrical about the mean, bell-shaped, mean/median and mode are identical, most scores clustered around the mean, relatively rare observations at the extremes, represents the ideal world for data analysis | What are the properties of normality? |
| You should make sure that your data roughly fits a bell curve shape before running certain statistical tests or regression. The tests that require normally distributed data incluce: Independent samples t-test | What is the assumption of normality? |
| If the standard deviation is high and the sample size is small <20 | When is normality a problem? |
| If we sample parameter estimates from a population, then as the sample size increase, thedistribution of those parameters becomes increasingly normal. | What is the Central Limit Theorem? |
| Analyse -> Descriptive Statistics -> Explore -> Drag variable into dependent list -> Display, select plots -> Plots -> NORMALITY plots with tests | What is the method of checking normality - objective? |
| Analyse -> Descriptive statistics -> Explore -> drag variable into dependent list -> display, select plots -> plots -> tick HISTOGRAM and also Normality plots with tests | What is the method of checking normality - subjective? |
| Use non-parametric tests / transformation of scores (Log 10, square root, Inverse, Reverse Scoring) | How do you deal with issues of normality? |
| Reduces positive skew | What does the transformation Log 10 do? |
| Reduces positive skew | What does the transformation, Square root, do? |
| Reduces positive skew | What does the transformation, Inverse, do? |
| Helps deal with negative skew | What does the transformation, Reverse Scoring, do? |
| Transform -> Compute Variable -> Name target variable -> Select SQUARE ROOT function -> Drag variable into task box | How do you transform variables? |
| Kolmogorov-Smirnov & Shapiro-Wilk | What are two tests of normality? |
| Historgram and Q-Q plot | What are two plots we can use to check normality? |
| Log10, square root, inverse and reverse scoring | Name 4 transformations |
| Testing the 'sameness'. Parametric tests assume that sampels are obtained from populations of equal variances. This means that the variability of scores for each of the groups need to be similar to each other to be comparable. SPSS performs Levene's test for equality of variances as part of the explore, t-test, and analysis of variance analyses. | What is homogeneity of variance? |
| Analyse -> Descriptive Stats -> Explore -> Drag IV to dependent list and DV to Factor list -> Plots, deselcft everything but tick POWER ESTIMATION | How do you conduct a Levene's test? |
| You are hoping to find that the test is NOT significant (i.e., a significance level greater than .05) | What does the Levene's output mean? |
| The variances for the two groups are not equal | What does it mean if you obtain a significance value of less than .05 on the Levene's test? |
| Research Question -> Null Hypothesis -> Alternative Hypothesis | What are the three components of quantitative research? |
| When the hypothesis is non-directional | What is a two-tailed test? |
| No matter what the distribution of the population is, the shape of the sampling distribution will approach normality as the sample size increases | What does the central limit theorem tell us? |
| The variance is not equal across groups | What is the null hypothesis for the assumption of homogeneity of variance? |
| Levene's Test | What is the name of the test used to examine homogeneity of variance? |
| Kolmogorov-Smirnov & Shapiro-Wilk | What are other normality tests called? |
| The assumption has been violated. You do NOT want a significant difference in tests of homogeneity of variance. If a violation occurs, it is likely that conducting the non-parametric equivalent of the analysis is more appropriate | What does a p-value of less than .05 mean for the homogeneity of variance? |
| Sig value <.05. square or cube the data | How can we identify the number required for power transformation (i.e., the transformation that is used when the assumption of homogeneity of variance is violated) |
| Null hypothesis | Which hypothesis (null or alternative) is always about lack of any effect or in other words no difference? |
| Alpha is the significance level. The probability that you will make the mistake of rejecting the null hypothesis when in fact it is true. The p-value measures the probability of getting a more extreme value than the one you got from the experience. If the p-value is greater than alpha, you accept the null hypothesis | What is the relationship between alpha (or significance level) and p-value |
| Incorrect rejection of a true null hypothesis. False positive. Leads one to conclude that a supposed IV effect, difference between groups, or a relationship exists when in fact it doesn't | What is a type 1 error? |
| Failure to reject a false null hypothesis. False negative. Failure to find and report a relationship when one actually exists. Having an adequate sample size is important to avoid this type of error occurring | What is a type 2 error? |
| Yes. The larger the sample size, the more likely a study will find a significant relationship exists (if one does). Random error is reduce with larger sample sizes | Is p-value affected by sample size? |
| Data view -> each row represents a case and each column represents a variable. Variable view -> contains descriptions of the attributes of each variable in the data file. | What is the difference between data view and variable view in SPSS? |
| Comparison of mean scores of two different groups of people or conditions | What is an independent samples t-test? |
| Comparison of mean score for the same group of people on two different occasions. | What is paired-samples t-test? |
| Analyse -> Compare means -> Independent Samples T-Test -> Drag continutous (DV) variable into test variable and categorical (IV) variable into grouping variable -> Define groups -> enter 1 for gp 1 and 2 for gp 2 -> Continue | How do you run an independent samples t-test? |
| Analyse -> compare means -> paired samples T-test | How do you run a paired-samples t-test? |
| Analyse -> compare means -> one sample t-test | How do you run a one-sample t-test? |
| To compare two independent groups of observations or measurements on a single characteristic. The test is the between-subjects analog and used for repeated measurement or matched observations | When is it appropriate to use independent t-test? |
| If you want to test a change or difference in means between two related groups, but not at the same time. | When is it appropriate to use dependent t-test? |
| Sample size minus 1 | How do you calculate df for one sample t-tests? |
| Number of pairs minus 1 | How do you calculate df for paired-samples t-test? |
| Sample size minus 2 | How do you calculate df for indepdnent samples t-tests? |
| Frequencies | When analysing categorical data, we typically analyse? |
| The p-value tells you the statistical significance of the difference; the t-value measures the size of the difference relative to the variation in your sample data. T is simply the calculated difference represented in units of standard error. The greater the magnitude of T, the greater the evidence AGAINST the null hypothesis. The closer that T is to 0, the more likely there isn't a significant difference. | What is the relationship between the t-statistic and the p-value? |
| The numeric value we attach to different categories are arbitrary. E.g., in SPSS, if Male = 1 and Female = 2, the mean of 1.5 bares no relation to the coding systen used for biological sex | Why is the mean of a categorical variable meaningless? |
| More statistical power because they control for factors that cause variability between subjects. Fewer subjects required. Quicker & Cheaper. Assess an effect over time. | What are the benefits of a repeated measures design? |
| Order effects (scores can decrease over time due to fatigue). Additional experimental materials required. | What are the limitations of a repeated measures design? |
| Frequencies actually observed in the data | What are observed frequencies? |
| Frequencies you would expect if the null hypothesis were true | Whare are expected frequencies? |
| Used to compare the proportion of cases from a sample (observed values) to theoretical values or those obtained previously from data (expected value). A relatively large chi-square value suggests a large discrepancy between observed and expected frequencies | What is a test for goodness of fit? |
| Random samples of data points. Each observation needs to be independent from one another (you can't have a single participant completing the survey more than once and there should be no interference between participants). Data being analysed needs to be categorical - therefore, no need for tests of normality or Levene's test of homogeneity of variance! | What are the assumptions for goodness of fit? |
| Equal probability, theory and previous data | What are 3 ways to derive expected frequencies? |
| Use the 'weight cases' function | When using aggregated data, what should you do to the data first? |
| That each cell has at least 5 observations | What is the additional assumption of the test of independence? |
| Standardised residuals | What statistic tells you what cells contributed the most to the significant results? |
| Used to determine whether two categorical variables are related. It is NOT a test of differences but is a test of relationships. It assumes that the data is nominal so these tests are non-directional | What is a chi-square teset of independence? |
| Random samples of data points. Each observation needs to be independent from one another. Data needs to be categorical. Each cell has at least 5 observations | Assumptions of a chi-square test of independence |
| The df is the number of cells (categories) -1 | How does SPSS calculate df in a goodness of fit test? |
| df = (r-1)(c-1) r=rows c=columns | How does SPSS calculate df in an independence test? |
| They can help tease out why the significant outcome occurred. SR >+2 or <-2 make notable contribution to the significant chi-square. Minus SR = fewer than expected and thus a noteworthy contributor. Plus SR = more than expected and thus a noteworthy contributor | How can we use standardised residuals to determine which cell(s) was (were) responsible for a significant chi-square statistic |
| The chi square goodness of fit test tests if the observed data fits a certain distribution or not. The chi square test of independence tests if the observed data does not significantly differ than what is expected under the independence assumption. The null hypotheses for the fitness test is that observed data follow the preassumed distribution while for test of independence the null is that the two categorical variables are independent | The difference between chi-square test for goodness of fit and chi-square test for independence |
