Regression Analysis Quiz

Predicts a dependent variable using two or more independent variables simultaneously.

The most appropriate line showing the relationship between dependent and independent variables.

The difference between the observed value and the true value (often unobserved).

<p>Variability in a DV uniquely explained by specific IV(s) in multiple regression, distinct from Pearson's where unique variance isn't assessed. Similar that they both are between 0-1.</p> Signup and view all the answers

<p>Variability in a DV explained by multiple IV's simultaneously in both multiple regression and Pearson's correlation.</p> Signup and view all the answers

<p>data = model + error</p> Signup and view all the answers

<p>Change in DV for one unit change in IV, holding other IV's constant - shared variance excluded.</p> Signup and view all the answers

<p>Difference between raw score and the mean score.</p> Signup and view all the answers

<p>Difference of predicted score from the mean (want to increase this).</p> Signup and view all the answers

<p>The error - difference between raw score and predicted score. Want to reduce this.</p> Signup and view all the answers

<p>Represents the proportion of variance in the DV that is explained by the IV in the model. Ranges from 0 (not explained) to 1 (explains all variability).</p> Signup and view all the answers

<p>The slope of the regression line reflecting the change in the DV from one unit change in the iV, whilst holding all other variables constant.</p> Signup and view all the answers

<p>The slopes of the regression line in standard deviation units (generally -1 to +1), making it comparable with other standardised coefficients. The change in SD units, for one SD change in IV, holding all other variables constant.</p> Signup and view all the answers

<p>Correlation between the predictor and outcome variable with variance shared between other predictors controlled in the predictor variable only.</p> Signup and view all the answers

<p>Correlation between a predictor variable and outcome variable while removing the shared variance with other predictors.</p> Signup and view all the answers

<p>It's an outlier score, that is unusual on the IV.</p> Signup and view all the answers

<p>Most unstable. Unusual on IV and DV, use Cooks distance or gap measure to identify.</p> Signup and view all the answers

<p>Unusual on IV, can use Mahanobalis or Leverage to identify.</p> Signup and view all the answers

<p>The relation between IV and the mean of DV is linear.</p> Signup and view all the answers

<p>The variance of redial is the same of any of IV.</p> Signup and view all the answers

<p>For any fixed value of IV, DV is normally distributed.</p> Signup and view all the answers

<p>That the association between predictors are independent.</p> Signup and view all the answers

<p>Both used to measure multicollinearity, tolerance is 0 (redundant) - 1 (independent). VIF is 10 redundant.</p> Signup and view all the answers

<p>To predict a DV using two ore more IV's. The IV's have equal importance to explanation.</p> Signup and view all the answers

<p>We determine what happens based on theory. Entered based on theoretical importance or control Eg. casual or exclusion. Change is obtained at each step.</p> Signup and view all the answers

<p>Not based on theory (not recommended), based on the size of the correlation. Largest correlation is entered first.</p> Signup and view all the answers

<p>Mediated regression uses an indirect effect, where the IV influences the DV by a pathway through a second IV (mediator).</p> Signup and view all the answers

<p>The IV should precede the mediator in time, and the mediator should preceded the DV.</p> Signup and view all the answers

<p>Two or more parallel mediator, that need each have an indirect pathway association.</p> Signup and view all the answers

<p>Path A, B, C, C' and a*b significant. Path C no longer required to be significant.</p> Signup and view all the answers

<p>a*b</p> Signup and view all the answers

<p>Test the significance of the mediated pathway, uses the 95% CI , sampling distribution. If contains 0 not statically significant.</p> Signup and view all the answers

<p>No longer used for significant testing for a mediated regression. need a high number and too conservative.</p> Signup and view all the answers

<p>The influence of one IV on DV changes based on score on second IV (moderator). The moderator is conditional and influences the direction or strength.</p> Signup and view all the answers

<p>The predictors each add variance to the explanation of the DV, so each predictor is independent, so additive influence on the outcome.</p> Signup and view all the answers

<p>The interaction between X*M.</p> Signup and view all the answers

<p>y=b0+b1X_b2M+b3(X*M)</p> Signup and view all the answers

<p>We use the unstandardised slope coefficient.</p> Signup and view all the answers

<p>The influence of each predictor is independent to other predictors. Lines are parallel.</p> Signup and view all the answers

<p>The effect of one predictor is dependent or conditional on the other predictor (slopes are different)</p> Signup and view all the answers

<p>It is used to produce a figure to describe the association. Mean and +/- 1 SD, to create low, mod, high. Can also use variable centring, produces a mean of 0.</p> Signup and view all the answers

<p>Old test, for describing the association, uses multiple levels of continuous moderator variables, more exact.</p> Signup and view all the answers

<p>Power, minimum 150 participants</p> Signup and view all the answers

<p>The covariate must have the same effect at each level of moderator variable, so does not produce a conditional effect itself.</p> Signup and view all the answers

<p>Tests whether two group means are significantly different.</p> Signup and view all the answers

<p>The ratio of the model to its error - between groups divided by within groups.</p> Signup and view all the answers

<p>The variance explained by our model?</p> Signup and view all the answers

<p>The variance not explained by our model (error).</p> Signup and view all the answers

<p>Test for an overall experimental effect, the difference lies somewhere.</p> Signup and view all the answers

<p>Between groups - independent groups (more error) Repeated measures (participants take part in all conditions) can control for error. Mixed: combination of repeated and independent.</p> Signup and view all the answers

<p>The influence of IV without regard for other IV's in the analysis.</p> Signup and view all the answers

<p>The influence of one IV on score of DV conditional on other IVs.</p> Signup and view all the answers

<p>The effect of one IV, differing at level of second IV.</p> Signup and view all the answers

<p>Overall effect of each IV of interest ignoring the other IV's.</p> Signup and view all the answers

<p>Specific set of cell means at different level of other IV's: eg. drugs: male/female (in univariate table)</p> Signup and view all the answers

<p>Both describe linear relationships between two variables, with range of 0 to 1. Bivariate is based on all variance in IV and DV. Semi-partial correlation shows how much unique variance is added to explanation when other IVs are controlled.</p> Signup and view all the answers

<p>How different the actual score is from the predicted score within the regression model. Whether there is an unusual score on the DV. More likely to be unusually high or low on DV. Between +/- 3 no problem - look at plots</p> Signup and view all the answers

<p>Measure of discrepancy, unusual score on IV and DV, occur at top or bottom of distribution only. Uses studenised residual and measure of leverage to give global measure of influence. Can use the gap method of cooks distance, to identify larger distances.</p> Signup and view all the answers

<p>It's a non-parametric sampling technique, generating multiple samples to generate a 95% CI for mediated effect. If contain 0, effect of mediated pathway is statistically significant.</p> Signup and view all the answers

<p>Size of factor loadings for individual items change with rotation. Groups with related items move closer to principal axis, increasing loading on one factor and reducing loadings on other factor/s. Rotated factors attempt to maximise the difference between factors.</p> Signup and view all the answers

<p>Rotations used to produce more coherent factors which have different items loading on different factors. Orthogonal used when factors independent, oblique solution used when factors not independent. Factor matrix used when single factor solution expected.</p> Signup and view all the answers

<p>Cronbach alpha assesses internal consistency of items on each factor, so will determine if factor is internally reliable. Need to have coefficient of .70 or better to use as a scale measure of a specific factor in research.</p> Signup and view all the answers

<p>When covariate is associated with DV &amp; not IV. – Used to reduce error. – Effects of IQ removed from problem solving, when effects of different amounts of alcohol intake on problem solving are to be evaluated</p> Signup and view all the answers

<p>Assumption of ANCOVA. Association between covariate &amp; DV is same for each group, i.e., slopes are same.</p> Signup and view all the answers

<p>Unconditional effects – Influence of each IV independent of one another, so additive – E.g., standard regression analysis; main effects only significant in ANOVA • Conditional effects: – Influence of one IV on DV influenced by score on additional IV, so influence of IVs not independent of one another – e.g., Significant moderator term in regression – e.g., Significant interaction in ANOVA/ANCOVA</p> Signup and view all the answers

<p>Balanced design: Equal number of participants in each group Unbalanced: when there are different numbers in each group. Effects can be correlated (hard isolate independent effects).</p> Signup and view all the answers

<p>Test determines whether variances &amp; covariance matrices across groups on repeated measures are same. Compares differences between group 1 &amp; 2 with differences between groups 1 &amp; 3 and 2 &amp; 3 across these measures. If same test is non-significant, sphericity is assumed.</p> <p>Not conducted if study has only two levels of repeated factor. Nothing to compare as only two groups.</p> Signup and view all the answers

<p>Tells us about adequacy of overall items in solution. Measures scores range from 0 to 1. Closer to one better the solution. Want to be &gt; .70</p> Signup and view all the answers

<p>Is a factor/PCA solution viable. If significant shows there are groups of independent or semi-independent items, so FA can be conducted. If not significant, identity matrix, so all items assessing same thing. PCA/FA should not be conducted.</p> Signup and view all the answers

<p>Unconditional effects consider the influence of each IV independent of the others, while conditional effects consider the influence of one IV influenced by another IV.</p> Signup and view all the answers

<p>It provides more accurate estimates of the population means.</p> Signup and view all the answers

<p>To determine if the covariance matrices across groups are equal.</p> Signup and view all the answers

<p>To measure the adequacy of the overall items in the solution.</p> Signup and view all the answers

<p>To test for the viability of a factor/PCA solution.</p> Signup and view all the answers

<p>The estimates of the population means are less accurate.</p> Signup and view all the answers

<p>The influence of one IV is influenced by another IV.</p> Signup and view all the answers

<p>To predict a dependent variable using multiple independent variables</p> Signup and view all the answers

<p>When there are only two levels of the repeated factor.</p> Signup and view all the answers

<p>It is a unique aspect of multiple regression</p> Signup and view all the answers

<p>The standardized slope is the unstandardized coefficient divided by the standard deviation of the independent variable</p> Signup and view all the answers

<p>To partition the total variance into explainable and unexplainable components</p> Signup and view all the answers

<p>SS total is the sum of SS regression and SS residual</p> Signup and view all the answers

<p>0 to 1</p> Signup and view all the answers

<p>To make the regression coefficient comparable across different variables</p> Signup and view all the answers

<p>To measure the proportion of variance explained by each independent variable</p> Signup and view all the answers

<p>Multicollinearity</p> Signup and view all the answers

<p>The order of variable entry</p> Signup and view all the answers

<p>a*b</p> Signup and view all the answers

<p>To test the significance of the mediated pathway</p> Signup and view all the answers

<p>The slopes are different</p> Signup and view all the answers

<p>To describe the association between the predictor and outcome variable</p> Signup and view all the answers

<p>Power</p> Signup and view all the answers

<p>A discrepancy score is unusual on the IV and DV, while an influential score is unusual on the IV</p> Signup and view all the answers

<p>Two or more mediators with indirect pathways</p> Signup and view all the answers

<p>b3</p> Signup and view all the answers

<p>The covariate is associated with the dependent variable and not the independent variable</p> Signup and view all the answers

<p>The overall effect of each independent variable of interest ignoring the other independent variables</p> Signup and view all the answers

<p>To generate a 95% confidence interval for the mediated effect</p> Signup and view all the answers

<p>Varimax rotation attempts to maximise the difference between factors, whereas unrotated factor matrix does not</p> Signup and view all the answers

<p>To determine the internal consistency of the items on each factor</p> Signup and view all the answers

<p>The effect of one independent variable differs at different levels of the second independent variable</p> Signup and view all the answers

<p>To determine the influence of an individual data point on the regression model</p> Signup and view all the answers

<p>Bivariate correlation shows the shared variance, whereas semi-partial correlation shows the unique variance added to the explanation</p> Signup and view all the answers

<p>To test the overall effect of the independent variables</p> Signup and view all the answers

<p>The assumption that the covariate has the same effect at each level of the moderator variable</p> Signup and view all the answers

<p>To predict a dependent variable using multiple independent variables</p> Signup and view all the answers

<p>Unique variance is the variability explained by a single independent variable, while shared variance is the variability explained by multiple independent variables</p> Signup and view all the answers

<p>To decompose the total variance into components</p> Signup and view all the answers

<p>The change in the dependent variable for one standard deviation change in the independent variable, holding other variables constant</p> Signup and view all the answers

<p>To represent the proportion of variance in the dependent variable explained by the independent variables</p> Signup and view all the answers

<p>The unstandardised coefficient is the slope of the regression line in original units, while the standardised coefficient is the slope of the regression line in standard deviation units</p> Signup and view all the answers

<p>The proportion of variance in the dependent variable explained by a specific independent variable</p> Signup and view all the answers

<p>When the covariate is associated with the DV and not the IV</p> Signup and view all the answers

<p>The overall effect of each IV of interest ignoring the other IVs</p> Signup and view all the answers

<p>The effect of one IV differing at the level of the second IV</p> Signup and view all the answers

<p>A measure of discrepancy, unusual score on IV and DV</p> Signup and view all the answers

<p>To generate multiple samples to generate a 95% CI for mediated effect</p> Signup and view all the answers

<p>The size of factor loadings for individual items change with rotation</p> Signup and view all the answers

<p>To determine if the factor is internally reliable</p> Signup and view all the answers

<p>The assumption that the covariate has the same effect at each level of moderator variable</p> Signup and view all the answers

<p>To test for an overall experimental effect</p> Signup and view all the answers

<p>Bivariate correlation describes linear relationships between two variables, while semi-partial correlation shows how much unique variance is added to explanation when other IVs are controlled</p> Signup and view all the answers

<p>Mediated regression</p> Signup and view all the answers

<p>To test the significance of the mediated pathway</p> Signup and view all the answers

<p>Pick a point technique</p> Signup and view all the answers

<p>Assumption of homoscedasticity</p> Signup and view all the answers

<p>Partial correlation</p> Signup and view all the answers

<p>Sobel test</p> Signup and view all the answers

<p>Moderation</p> Signup and view all the answers

<p>b3</p> Signup and view all the answers

<p>150</p> Signup and view all the answers

<p>Interactive model</p> Signup and view all the answers

<p>Unconditional effects consider the influence of one independent variable on the dependent variable, while conditional effects consider the influence of multiple independent variables.</p> Signup and view all the answers

<p>The effects of the independent variables are correlated, making it difficult to isolate independent effects.</p> Signup and view all the answers

<p>To determine if the variances and covariance matrices are equal across groups.</p> Signup and view all the answers

<p>The adequacy of the overall items in the solution.</p> Signup and view all the answers

<p>To determine if the solution is viable.</p> Signup and view all the answers

<p>The number of participants in each group.</p> Signup and view all the answers

<p>Sphericity is assumed.</p> Signup and view all the answers

<p>When there are only two levels of the repeated factor.</p> Signup and view all the answers