PSY4062
Interpreting skew/kurtosis testing SPSS
0 = no skew or kurtosis - Convert to a z-score (use cut off of 1.96 which = p <.05) to determine if skew/kurtosis is statistically significant.- Do this by taking stat from SPSS and dividing by standard error EXAMPLE; Stat / Standard Error -.003 / .086 = -0.03 this is not greater than 1.96, or less than -1.96 therefore no skew/kurtosis
Greenhouse-Geisser
A conservative adjustment to the degrees of freedom in repeated measures analysis of variance to correct for violation of the assumption of sphericity.
Homoscedasticity
A regression in which the variances in y for the values of x are equal or close to equal
Heteroscedasticity
A regression in which the variances in y for the values of x are not equal
Multicollinearity
A situation in which several independent variables are highly correlated with each other. This characteristic can result in difficulty in estimating separate or independent regression coefficients for the correlated variables.
Ordinal data
A statistical data type that exists on an arbitrary numerical scale where the exact numerical value has no significance other than to rank a set of data points. Deals with the order or position of items such as words, letters, symbols or numbers arranged in a hierarchical order. Quantitative assessment cannot be made.
Bonferroni correction
A way of creating a more stringent criterion for accepting F as significant.
Independent measures
AKA - between subjects
Repeated measures
AKA - within subjects
You don't need to interpret main effects if...
An interaction involving that variable is significant
Bonferroni post hoc tests
Break down the main effect of an IV and can be interpreted as if one IV was the only predictor in the model
Mixed measures
Combo of between and within subjects
Helmert contrast
Compares each category against all subsequent categories.
Simple contrast
Compares one group mean to another.
Repeated contrast
Contrast that moves through the groups, comparing each group to the one before.
Shapiro-Wilk test
Dedicated test for normality. If Sig < .05 then the assumption of normality has been violated.
Main effect
In a factorial design, the overall effect of one independent variable on the dependent variable, averaging over the levels of the other independent variable.
F-statistics are not _______, so they do not control the _____
Independent, Type I error rate
Simple effects analysis
Looks at the effect of one independent variable at individual levels of the other independent variable.
How is F computed?
Mean squares for the effect divided by the residual mean squares
Leptopkurtic/Positive Kurtosis
More values in the distribution tails and more values close to the mean (i.e. sharply peaked with heavy tails). A tall and slim bell curve
Normality of residuals
Normality is the assumption that the underlying residuals are normally distributed, or approximately so
Repeated-measures (related) factorial design
Several independent variables or predictors have been measured, but the same entities have been used in all conditions.
Mixed design
Several independent variables or predictors have been measured; sme have been measured with different entities, whereas others used the same entities.
What do contrasts do?
Tell us where group differences lie
Independent factorial design
There are several independent variables or predictors, and each has been measured using different participants.
Issues with outliers
They bias the mean and inflate the standard deviation
Which correlation coefficient explains more of the variance? a. .70 b. .60 c. 0 d. -.80
To determine the proportion of variability accounted for, you need to calculate the coefficient of determination by squaring the correlation coefficient: r r2 .70 .49 .60 .36 -.80 .64 0 0 Thus, a correlation coefficient of -.8 explains more variability (i.e., 64%) than the other correlation coefficients. The correct answer is: d) -.80
Dealing with outliers
Truncate - drop them Winsorization - change outlier to next highest score +1 Robustness - analyse data with a robust procedure
Two way independent measures ANOVA
When there's two independent variables. For example, a two-way ANOVA allows a company to compare worker productivity based on two independent variables, such as department and gender.
Durbin-Watson statistic
a number that tests for autocorrelation in the residuals from a statistical regression analysis. Always has a value ranging between 0 and 4. A value of 2.0 indicates there is no autocorrelation detected in the sample. Values from 0 to less than 2 point to positive autocorrelation and values from 2 to 4 means negative autocorrelation.
Interaction effect
a result from a factorial design, in which the difference in the levels of one independent variable changes, depending on the level of the other independent variable; a difference in differences. On a graph - parallel lines means NO interaction effect
Factorial design
a study that explores the impact of two or more factors (independent variables) on a particular dependent variable. F ratio for every factor, and for every interaction
Ratio data
a type of numerical data in which the difference between numbers is significant, but there is a fixed non-arbitrary zero point associated with the data
The correlation between Y and X1 is .80. The correlation between Y and X2 is .90. A multiple regression using X1 and X2 to predict Y could explain no less than ______ and no more than _____ of the variance in Y. a) .81, 1.00 b) .72, 1.53 c) .72, .81 d) .81, .81
a) .81, 1.00
Which of the following is considered the minimum acceptable level for power? a. 0.80 b. 0.05 c. 0.95 d. 1.00
a. 0.80
For a 2x3 independent-measures ANOVA, there are ____ F ratios, and ____ error terms. a. 3, 1 b. 3, 2 c. 6, 5 d. 2, 3
a. 3, 1 '2x3' indicates that this is a two-way design, with 2 levels of factor A, and 3 levels of factor B. As a two-way ANOVA, it will have 3 F-ratios: * One for the main effect of factor A * One for the main effect of factor B * One for the interaction between factor A and factor BSince it is a two-way independent-measures ANOVA, there will be just one error term (i.e., the residual variance - that is, the variance not explained by the effects of factor A, factor B, and the AxB interaction)
An experiment was conducted to examine the effectiveness of different treatments for social anxiety. There were 3 treatment conditions: i) control (no treatment); ii) exposure therapy; and iii) medication only. Participants were randomly assigned to each treatment condition, and their levels of anxiety were measured both before treatment, and six weeks after treatment. A two-way mixed model ANOVA was run to analyse Anxiety Levels as a function of Treatment Condition (control vs. exposure therapy vs. medication; Factor A), and Time (before vs. after treatment; Factor B). Assuming that there were 12 participants in each Treatment Condition, and each participant has their Anxiety Level measured twice, how many rows of data would SPSS expect? a. 36 b. 24 c. 72 d. 12
a. 36
An AxB design with 2 levels of factor A and 6 levels of factor B, has a significant effect of factor B. How many orthogonal comparisons can be performed to determine where the differences are actually occurring within the significant F ratio? a. 5 b. 3 c. 4 d. 11
a. 5 You are told that the main effect of factor B is significant. Since there are 6 levels of factor B, that means that that the main effect of factor B compares 6 group means. The number of possible orthogonal comparisons is equal to the number of groups minus one. Thus, for factor B, there are 6 - 1 = 5 possible orthogonal comparisons.
Which of the following is required before running a simple effects analysis? a. A significant interaction effect b. At least three factors c. Orthogonal contrast weights d. Specific hypotheses
a. A significant interaction effect
If you have a curvilinear relationship, then: a. It is not appropriate to use Pearson's correlation because it assumes a linear relationship between variables. b. You can use Pearson's correlation; you just need to remember that a curve indicates that the variables are not linearly related. c. Pearson's correlation can be used in the same way as it is for linear relationships. d. Transforming the data won't help.
a. It is not appropriate to use Pearson's correlation because it assumes a linear relationship between variables.
What is the formula for calculating the F ratio for a regression analysis? a. MSRegression / MSResidual b. MSRegression / MSTotal c. MSTotal / MSRegerssion d. MSResidual / MSTotal
a. MSRegression / MSResidual
In a regression analysis, the tolerance statistic tells you about which statistical assumption? a. Multicollinearity b. Heterscedasticity of residuals c. Independence of errors d. Normality
a. Multicollinearity
A two-way ANOVA reveals a significant interaction between factors A and B on the DV. What is the most appropriate approach for interpreting this interaction? a. Perform simple effects analysis b. No further interpretation is necessary c. Graph the interaction d. Examine the main effects in detail
a. Perform simple effects analysis
In hierarchical multiple regression, the change in _______ is used to test whether adding blocks of variables increases the amount of variance explained by the model. a. R2 b. sr2 c. R d. Adjusted R2
a. R2
The relationship between two variables partialling out the effect that a third variable has on one of those variables can be expressed using a: a. Semi-partial correlation b. Partial correlation c. Point-biserial correlation d. Bivariate correlation
a. Semi-partial correlation
____________ examine the effects of one IV at one level of the other IV, then repeats the process for all other levels. a. Simple effects analysis b. Interaction effects c. Planned comparisons d. Post-hoc tests
a. Simple effects analysis
What is a potential advantage of a repeated-measures design, compared to an independent-measures design? a. Smaller sample size required b. Only need to measure each participant once c. It eliminates confounding variables d. Reduced risk of attrition
a. Smaller sample size required
A market researcher wants to identify which variables (out of a set of 10) predict how much customers spend in an online store. She wants to end up with a model where all of the predictors contribute to predicting the outcome variable. There is no prior information to guide her choice of variables. Which regression method is most appropriate to achieve this? a. Stepwise hierarchical multiple regression b. Simple regression c. Standard multiple regression d. Standard Hierarchical regression
a. Stepwise hierarchical multiple regression
The Durbin-Watson statistic informs us about the assumption of independence of errors. If an analysis produces a Durbin-Watson value of 0.03, what should we conclude about the independence of errors? a. The assumption is likely violated as the value is < 1 b. The assumption is likely violated as the value is > 0 c. The assumption is not likely violated as the value is < 2 d. The assumption is not likely violated at the value is close to 0
a. The assumption is likely violated as the value is < 1
A researcher wants to find out if kurtosis is a problem in her data. She generates a raw kurtosis statistic of 2.05. Which is the most correct statement about kurtosis in the data? a. There is not enough information to assess kurtosis b. The data have statistically significant negative kurtosis c. The data have statistically significant positive kurtosis d. The data do not have statistically significant kurtosis
a. There is not enough information to assess kurtosis
A standard multiple regression analysis was performed to determine whether 185 students' exam scores (Exam Scores) could be predicted by the number of classes they attended (Attendance), the amount of time they spent revising (Revision Time), and their level of exam anxiety (Anxiety). Which of the following statements about this standard multiple regression analysis is FALSE? a. There was one independent variable b. There were 185 rows of data in the SPSS datasheet c. There were four columns of data in the SPSS datasheet d. There was one dependent variable
a. There was one independent variable
In a study of depression, you measured depression scores (a continuous, normally distributed variable) at baseline, 1 month, 6 months, and 12 months. What statistical test will best tell you whether or not two separate treatments for depression (administered to different participants) had different effects over time? a. Two-way mixed-model ANOVA b. One-way repeated-measures ANOVA c. Two-way independent-measures ANOVA d. Two-way repeated-measures ANOVA
a. Two-way mixed-model ANOVA
Given a significant F comparing k treatments, how many orthogonal contrasts will be significant? a. at least one b. one c. k d. k-1
a. at least one
The results from a two-way analysis of variance show a significant main effect for factor A and a significant main effect for factor B. a. you cannot make any conclusion about the significance of the interaction b. there is a probably a significant interaction c. there must be a significant interaction d. the interaction cannot be significant
a. you cannot make any conclusion about the significance of the interaction
Factorial ANOVA
an analysis of variance involving two or more independent variables or predictors. better reflects the real world = better ecological validity More efficient
Huynh-Feldt estimate
an estimate of the departure from sphericity. The maximum value is 1 (the data completely meet the assumption of sphericity). Values below this indicate departures from sphericity and are used to correct the degrees of freedom associated with the corresponding F-ratios by multiplying them by the value of the estimate. It is less conservative than the Greenhouse-Geisser estimate, but some say it is too liberal.
Levene's test
assesses the assumption of homogeneity of variance (i.e., that each group's scores have approximately equal variance). p = >.05 is non-significant, i.e. homogeneity of variance is assumed
Kolmogorov-Smirnov test
assesses the assumption of normality Sig < 0.05 means variable is NOT normally distributed
Mauchly's test
assesses the assumption of sphericity (i.e., if the difference between each group is calculated, that the variances of the difference scores are approximately the same
Linearity
assumption that the best way to describe a pattern of data is using a straight line
The correlation between Y and X1 is .50 and the correlation between Y and X2 is .30. What value would the correlation between X1 and X2 need to be such that a multiple regression using BOTH X1 and X2 to predict Y explains .34 of the variance of Y? Select one: a. -1 b. 0 c. 1 d. Impossible to tell
b. 0
Of the following Durbin-Watson values for a particular regression analysis, which would indicate the least concern regarding independence of errors? a. 4 b. 2 c. 3 d. 1
b. 2
Imagine you run a three-way independent measures ANOVA. How many two-way interaction effects are there? a. 4 b. 3 c. 2 d. 1
b. 3
A factorial ANOVA is performed on a fully independent measures AxB design, with 2 levels of factor A, and 3 levels of factor B. How many F ratios are there? a. 2 b. 3 c. 6 d. 4
b. 3 Because there are two factors, this is a two-way independent-measures ANOVA. Two-way ANOVAs have 3 F-ratios: * One for the main effect of factor A * One for the main effect of factor B * One for the interaction effect between factor A and factor B
n experiment was conducted to examine the effectiveness of different treatments for social anxiety. There were 3 treatment conditions: i) control (no treatment); ii) exposure therapy; and iii) medication only. Participants were randomly assigned to each treatment condition, and their levels of anxiety were measured both before treatment, and six weeks after treatment. Imagine participants had their anxiety measured at 5 time points instead of 2. How many columns of data would SPSS expect? a. 5 b. 6 c. 7 d. 4
b. 6
An experiment was conducted to examine the effectiveness of different treatments for social anxiety. There were 3 treatment conditions: i) control (no treatment); ii) exposure therapy; and iii) medication only. Participants were randomly assigned to each treatment condition, and their levels of anxiety were measured both before treatment, and six weeks after treatment. Imagine participants had their anxiety measured at 6 time points instead of 2. How many columns of data would SPSS expect? a. 4 b. 7 c. 6 d. 5
b. 7
For a regression model, which coefficient tells you how much variance the model explains in the outcome variable if the model had been derived from the population? a. F b. Adjusted R2 c. R d. R2
b. Adjusted R2
A researcher conducts a two-way independent measures ANOVA and finds a statistically significant interaction effect. How many statistically significant main effects are there likely to be? a. 2 b. Cannot tell from the information provided c. 0 d. 1
b. Cannot tell from the information provided
A researcher finds a significant interaction between sex (male and female) and diagnosis (schizophrenia, bipolar disorder, and controls) in terms of depression score (the dependent variable). Which of the following is an appropriate strategy for running a simple effects analysis? a. For each level of diagnosis, run a two-way ANOVA on sex b. For each level of sex, run a one-way ANOVA on diagnosis c. For each level of diagnosis, run a two-way ANOVA on depression d. For each level of sex, run a t-test on diagnosis
b. For each level of sex, run a one-way ANOVA on diagnosis
In general, independent-measures designs are ________________ than repeated-measures designs. a. More powerful b. Less powerful c. Comparable in power d. Impossible to predict
b. Less powerful
To determine how much additional variation is accounted for by the Block 2 variables in a hierarchical multiple regression, we look at the: a. ANOVA Table b. Model Summary R Square Change c. Model Summary F Change d. Model Summary Adjusted R Square
b. Model Summary R Square Change
In a multiple regression analysis in SPSS, the VIF statistic tells you about which statistical assumption? a. Normality b. Multicollinearity c. Homoscedasticity d. Independence of errors
b. Multicollinearity
A histogram in which high scores are most frequent (i.e. bars on the graph are highest on the right hand side) is said to be: a. Leptokurtic b. Negatively skewed c. Positively skewed d. Platykurtic
b. Negatively skewed
A researcher wants to understand the relationship between extroversion and self esteem controlling for the effect of sociability on both variables. Which correlation coefficient is most appropriate? a. Semi-partial correlation b. Partial correlation c. Zero-order correlation d. Standardised regression coefficient
b. Partial correlation
A two-way ANOVA reveals a significant interaction between factors A and B on the DV. What is the most appropriate approach for interpreting this interaction? a. Graph the interaction b. Perform simple effects analysis c. Examine the main effects in detail d. No further interpretation is necessary
b. Perform simple effects analysis
In hierarchical multiple regression, the change in _______ is used to test whether adding blocks of variables increases the amount of variance explained by the model. a. sr2 b. R2 c. Adjusted R2 d. R
b. R2
A researcher is examining the relationship between depression and adult attachment. However, they have reason to believe that childhood attachment influences both of these, and want to control this influence. Which of the following, if any, should they do? a. Run a semi-partial correlation between depression and adult attachment controlling for childhood attachment b. Run a partial correlation between depression and adult attachment controlling for childhood attachment c. Run a Spearman correlation between depression and adult attachment controlling for childhood attachment d. None of the above are possible
b. Run a partial correlation between depression and adult attachment controlling for childhood attachment
What is the formula for R2 in a regression analysis? a. SSTotal / SSRegression b. SSRegression / SSTotal c. SSResidual / SSTotal d. SSRegression / SSResidual
b. SSRegression / SSTotal
The relationship between two variables partialling out the effect that a third variable has on one of those variables can be expressed using a: a. Partial correlation b. Semi-partial correlation c. Bivariate correlation d. Point-biserial correlation
b. Semi-partial correlation
The results of a one-way repeated-measures ANOVA with four levels on the independent variable revealed a significance value for Mauchly's test of p = 0.048. What does this mean? a. The assumption of sphericity has been met. b. The assumption of sphericity has been violated. c. This value can be ignored because sphericity is not an issue in a one-way repeated-measures ANOVA design. d. That Tukey's test should be used.
b. The assumption of sphericity has been violated.
Which of the following does NOT apply when you have data that violate the assumption of sphericity? a. The Greenhouse-Geisser or Huynh-Feldt correction should be applied. b. The means are adjusted for any groups that are affected by the violation using estimates of sphericity. c. You can use multivariate test statistics (MANOVA) instead. d. The degrees of freedom are adjusted for any F-ratios affected by the violation using estimates of sphericity.
b. The means are adjusted for any groups that are affected by the violation using estimates of sphericity.
A researcher conducts a factorial ANOVA with two factors. Participants can only have data for one-level of each factor. What kind of design is this? a. Two-way mixed design ANOVA b. Two-way independent measures ANOVA c. Two-way within-subjects ANOVA d. Two-way repeated measures ANOVA
b. Two-way independent measures ANOVA
When are post hoc tests used? a. When there is a large Type I error b. When there are no specific hypotheses before the experiment c. When the familywise error is large d. None of the above
b. When there are no specific hypotheses before the experiment
The Kolmogorov-Smirnov test can be used to test: a. Whether group means differ. b. Whether scores are normally distributed. c. Whether scores are measured at the interval level. d. Whether group variances are equal.
b. Whether scores are normally distributed.
In stepwise hierarchical regression, ____________ starts with all possible predictors entered into the model. a. forward selection b. backward selection c. stepwise selection d. standard multiple regression
b. backward selection
In a two-way analysis of variance, a main effect is defined as the: Select one: a. difference between the largest treatment mean and the smallest treatment mean b. mean differences among the levels of one factor c. mean differences among all the treatment conditions d. mean difference between the two factors
b. mean differences among the levels of one factor
A researcher obtains a zero-order Pearson r of .40. This is best described as a(n) _________ effect size. a. small b. medium c. impossible d. large
b. medium
Statistical power is the probability of a. accepting the alternative hypothesis if the alternative hypothesis is false. b. rejecting the null hypothesis if the null hypothesis is false. c. rejecting the alternative hypothesis if the null hypothesis is true. d. rejecting the null hypothesis if the null hypothesis is true.
b. rejecting the null hypothesis if the null hypothesis is false.
For a 2x3 repeated-measures ANOVA, there are ____ F ratios, and ____ error terms. a. 3, 2 b. 2, 3 c. 3, 3 d. 6, 5
c. 3, 3
An experiment was conducted to examine the effectiveness of different treatments for social anxiety. There were 3 treatment conditions: i) control (no treatment); ii) exposure therapy; and iii) medication only. Participants were randomly assigned to each treatment condition, and their levels of anxiety were measured both before treatment, and six weeks after treatment. A two-way mixed model ANOVA was run to analyse Anxiety Levels as a function of Treatment Condition (control vs. exposure therapy vs. medication; Factor A), and Time (before vs. after treatment; Factor B). Assuming that there were 11 participants in each Treatment Condition, and each participant has their Anxiety Level measured twice, how many rows of data would SPSS expect? a. 11 b. 66 c. 33 d. 22
c. 33
In order to run a two-way independent-measures ANOVA using SPSS, which of the following should you select: a. Analyze -> General Linear Model -> Two-way ANOVA b. Analyze -> Descriptive Statistics -> Two-way ANOVA c. Analyze -> General Linear Model -> Univariate d. Analyze -> General Linear Model -> Multivariate
c. Analyze -> General Linear Model -> Univariate
To conduct a mixed model ANOVA using SPSS, you should select: a. Analyze ⇨ General Linear Model ⇨ Multivariate b. Analyze ⇨ Mixed Models ⇨ Linear c. Analyze ⇨ General Linear Model ⇨ Repeated Measures d. Analyze ⇨ General Linear Model ⇨ Univariate
c. Analyze ⇨ General Linear Model ⇨ Repeated Measures
I divide my study population into smokers, ex-smokers, and never-smokers. I want to compare years of schooling (a normally distributed variable) between the three groups. What test should I use? a. Paired-samples t-test b. One-way repeated-measures ANOVA c. One-way independent-measures ANOVA d. Two-way independent-measures ANOVA
c. One-way independent-measures ANOVA
Which correlation coefficient/s can tell you how much variability in the outcome is uniquely accounted for by that predictor? a. Zero-Order correlations b. Partial correlations c. Part correlations d. All of the above
c. Part correlations
A lecturer wants to predict her student's final mark (%) based on their mean quiz mark (%) and previous year's mark (%). Which analysis is most appropriate? a. Partial correlation b. Semi-partial correlation c. Standard multiple regression d. Two-way independent measures ANOVA
c. Standard multiple regression
A researcher runs a multiple regression analysis and calculates a Durbin-Watson statistic of 0.54. Which is the most correct interpretation? a. The assumption of multicollinearity is satisfied b. The assumption of multicollinearity is violated c. The assumption of independence of errors is violated d. The assumption of independence of errors is satisfied
c. The assumption of independence of errors is violated
A simple effects analysis is defined as: a. The difference between the main effect of factor A and the main effect of factor B b. The effect of one independent variable taken by itself c. The effect of one independent variable at a single level of the other independent variable d. Part of the interaction
c. The effect of one independent variable at a single level of the other independent variable
Which of the following is not a likely reason for a non-significant finding? a. The study was poorly designed or controlled b. The sample size was too small c. The effect size was too large d. There was really no difference between conditions
c. The effect size was too large
In a factorial ANOVA what is the overall effect of an independent variable on a dependent variable known as? a. The interaction effect b. The indirect effect c. The main effect d. The direct effect
c. The main effect
Which of the following is always an assumption of parametric statistical methods? a. The sample distribution is normally distributed b. The population distribution is normally distributed c. The sampling distribution is normally distributed d. None of the above
c. The sampling distribution is normally distributed
An experiment was done to look at whether there is an effect of the number of hours spent practising a musical instrument and gender on the level of musical ability. A sample of 30 (15 men and 15 women) participants who had never learnt to play a musical instrument before were recruited. Participants were randomly allocated to one of three groups that varied in the number of hours they would spend practising every day for 1 year (0 hours, 1 hours, 2 hours). Men and women were divided equally across groups. All participants had a one-hour lesson each week over the course of the year, after which their level of musical skill was measured on a 10-point scale ranging from 0 (you can't play for toffee) to 10 ('Are you Mozart reincarnated?'). Which of the following tests could we use to analyse these data? a. t-test b. Two-way repeated-measures ANOVA c. Two-way independent ANOVA d. Three-way mixed design ANOVA
c. Two-way independent ANOVA
An experiment was done to compare the effect of having a conversation via a hands-free mobile phone, having a conversation with an in-car passenger, and no distraction (baseline) on driving accuracy. Twenty participants from two different age groups (18-25 years and 26-40 years) took part. All participants in both age groups took part in all three conditions of the experiment (in counterbalanced order), and their driving accuracy was measured by a layperson who remained unaware of the experimental hypothesis. Which of the following would be the most appropriate method for analysing these data? a. Two-way independent ANOVA b. Two-way repeated-measures ANOVA c. Two-way mixed ANOVA d. One-way repeated-measures ANOVA
c. Two-way mixed ANOVA
Which of the following is LEAST useful in identifying the importance of predictors in a multiple regression analysis? a. Partial correlation coefficient b. Semi-partial correlation coefficient c. Unstandardised regression coefficient d. Standardised regression coefficient
c. Unstandardised regression coefficient
For k groups, ____ coefficients would be required to perform ____ orthogonal contrasts. a. k-1, k-1 b. k-1, k c. k, k-1 d. k, k
c. k, k-1
In multiple regression, when we speak of high correlations amongst the independent variables (i.e., predictors), we are speaking of: a. multiple correlation b. independence c. multicollinearity d. homoscedasticity
c. multicollinearity
A standard regression analysis is most appropriate when the research question involves _____________. a. non-linear trends b. association between two variables c. prediction d. group differences
c. prediction
Orthogonal contrasts
contrasts (i.e., planned comparisons) which are independent of one another. That is, contrasts which do not tap into the same difference twice - or, in other words, contrasts where there is no redundancy. The number of possible orthogonal contrasts is always equal to one less than the number of groups. So if there are k groups, then there are k - 1 possible orthogonal contrasts.
A Pearson correlation of -.30 is obtained between apple and orange sizes in orchards. It is not statistically significant. What can be concluded? a. Negative relationship b. 30% of the variance is explained c. Comparisons are meaningless d. Any linearity observed could be due to chance
d. Any linearity observed could be due to chance
Which of the following is not an assumption of a 3 x 3 mixed-design ANOVA? a. Homoscedasticity b. Sphericity c. Independence d. Heterogeneity of variance
d. Heterogeneity of variance
A researcher conducts a two-way mixed design ANOVA. Mauchly's test returns a p value of .04 and the Greenhouse-Geisser epsilon value is .92. In SPSS, which line from the 'test of within-subjects effects' table should be interpreted? a. Sphericity Assumed b. Lower-bound c. Greenhouse-Geisser d. Huynh-Feldt
d. Huynh-Feldt
If the correlation between education and income is .80, and between income and IQ is .60, then the correlation between IQ and education is _____? a. .20 b. .48 c. .60 d. Impossible to tell
d. Impossible to tell
All other things being held constant, adding more predictors into a regression model will ____________ the value of R2. a. Decrease b. Not change c. Invalidate d. Increase
d. Increase
A Pearson's correlation coefficient of .50 would produce a scatterplot in which the slope: a. Is vertical. b. Is downwards (from the bottom right corner to the top left corner of the graph). c. Is flat (horizontal). d. Is upwards (from the bottom left corner to the top right corner of the graph).
d. Is upwards (from the bottom left corner to the top right corner of the graph).
Which of the following is not corrected for by calculating Adjusted R2? a. Number of IVs b. Bias c. Sample Size d. Multicollinearity
d. Multicollinearity
Which of the following is not used to evaluate parametric test assumptions? a. Levene's test b. Kolmogorov-Smirnov test c. Mauchly's test d. None of the above
d. None of the above
After an ANOVA you need more analysis to find out which groups differ. When you did not generate specific hypotheses before the experiment use: a. Planned contrasts b. Bootstrapping c. t-tests d. Post hoc tests
d. Post hoc tests
The assumption of homogeneity of variance is met when: a. The variance across groups is proportional to the means of those groups. b. The variances in different groups are significantly different. c. The variance is the same as the interquartile range. d. The variances in different groups are approximately equal.
d. The variances in different groups are approximately equal.
A significant Pearson's correlation of -.71 was found between number of hours spent at work and energy levels in a sample of 300 participants. Which of the following conclusions can be drawn from this finding? a. Spending more time at work caused participants to have less energy. b. Amount of time spent at work accounted for 71% of the variance in energy levels. c. The estimate of the correlation will be imprecise. d. There was a strong negative relationship between the number of hours spent at work and energy levels.
d. There was a strong negative relationship between the number of hours spent at work and energy levels
A standard multiple regression analysis was performed to determine whether 185 students' exam scores (Exam Scores) could be predicted by the number of classes they attended (Attendance), the amount of time they spent revising (Revision Time), and their level of exam anxiety (Anxiety). Which of the following statements about this standard multiple regression analysis is TRUE? a. There were 180 rows of data in the SPSS datasheet b. There was one independent variable c. There were three columns of data in the SPSS datasheet d. There was one dependent variable
d. There was one dependent variable
If two variables are significantly correlated, r = .67, then: a. The variables are independent of one another. b. There is no unique variance. c. The relationship is weak. d. They share variance.
d. They share variance.
When is it not appropriate to conduct a simple effects analysis? a. When the data are not normally distributed b. When you do not have a specific hypothesis c. When none of the main effects are statistically significant d. When the interaction effect is not statistically significant
d. When the interaction effect is not statistically significant
Nominal Data
data of categories only. Data cannot be arranged in an ordering scheme. (Gender, Race, Religion)
Interval data
differences between values can be found, but is NO absolute ZERO. Examples: temperature, time
Platykurtic/Negative Kurtosis
flat curve, like the plateau of a mountain
Inter-item correlation
indication of the degree to which each item correlates with each other. Higher the value (or closer to 1) the more correlated they are.
Item-total stats
indication shows you the degree to which each item correlates with the total score, low values indicate that the item is measuring something different from the scale as a whole
Chronbach's Alpha
measure of internal consistency
If significant main effects are not qualified by an interaction then consult....
post-hoc tests
Mean squares
sum of squares divided by degrees of freedom
Negative skew
tail to the left
Positive skew
tail to the right
Mauchly's test
test of sphericity tests the null hypothesis that variances between level differences are equal <0.05 = assumption is violated then must look at the Epsilon values
Central Limit Theorem (CLT)
the distribution of sample averages tends to be normal regardless of the shape of the process distribution.
The F ratio
the ratio of between-groups variance to within-groups variance. Larger ratio = bigger effect = more likely to reject null hypothesis
Sums of squares
the sum of the deviation scores for a distribution