Psychological Statistics Exam (ANOVA) Questions With Multiple Choices Questions And Answers

What is a factor? - correct answers An independent or quasi-independent variable. Ex. location, language, medication, lighting level... What are levels? What letter is used to denote levels/conditions? - correct answers Levels are the individual conditions or values that make up a factor (conditions of the independent variable). Level is denoted by k. Ex. bright lighting, dim lighting, no lighting; two pills, one pill, no pills... What is the null hypothesis for an independent-measures single factor ANOVA (in symbols and words)? - correct answers H₀: μ₁ = μ₂ = μ₃ There are no significant differences between any of the means. What is the alternative hypothesis for an independent-measures single factor ANOVA (in words)? - correct answers There is at least one mean difference among the populations. What is a quasi-independent variable? - correct answers An independent variable that is NOT manipulated. Ex. age, weight, height... Determine the factor(s) and level(s): - Self-Deception - High deception, medium deception, low deception - correct answersSelf-deception is the factor. High, medium, and low deception are the levels of the factor (levels of self-deception). In a factorial ANOVA can you mix independent and repeated measures designs? - correct answersYes. What is the difference between an independent samples, factorial, and repeated-measures ANOVA? - correct answersIndependent samples are completely independent from each other. The samples are unrelated and use different individuals from each other. A factorial ANOVA involves multiple factors, or independent variables. Rather than just looking at three levels of the variable time, it can compare time and all its levels to type of therapy (and more). A repeated-measures ANOVA uses the same sample of participants for multiple conditions. They may be tested before, during, and after a treatment to look at how they changed during it. What does between-subjects mean? - correct answersIt looks at how much variability exists between the different conditions/levels/groups. Ex. if between-subjects variability is high between the no medication and medication conditions, that implies that the medication is having an effect. Does higher between-subjects or within-subjects variability lead to a larger (more significant) F-ratio? - correct answersBetween-subjects Explanation: Between-subjects variability refers to the difference between conditions/levels/groups. A big difference between conditions means that the treatment is probably having a significant effect, which means a bigger F-ratio. SStotal = 60 SSwithin = 40 What is SSbetween? - correct answersSSbetween = 20 Explanation: SStotal = SSbetween + SSwithin. In an independent-measures ANOVA, dftotal = 40. What is N? - correct answersN = 41 Explanation: dftotal = N - 1. You were given 2 out of 3 numbers so you can fill those in and then use algebra to solve the problem for the missing variable (N). dftotal = N - 1 40 = N - 1 Add 1 to each side 41 = N Approximate what the F-ratio would be if the null hypothesis is true. - correct answersF = 1.00 Explanation: The F number is a ratio, so a score of 1.00 means that the top and bottom (msbetween and mswithin) are exactly the same. This means that the variance between conditions is not difference from the random, unsystematic variance (which is really bad!) A significant F-ratio - correct answersThe bigger the better when it comes to F-ratios. A bigger F number means that the variance between conditions is high (which means that your treatments are different from each other). ex: 17.45 If dftotal = 39, how many conditions could the study have? Choose two. a) 4 b) 3 c) 6 d) 2 - correct answersa) 4 and d) 2 Explanation: dftotal = N - 1 39 = N - 1 40 = N Participants can't be decimals so it must be able to divide evenly. 4 x 10 = 40, so 4 groups would work fine. The same goes for 2 because 2 x 20 = 40. 3 and 6 do not divide evenly into 40. When looking up F-critical scores on the table, which two numbers do you use? Which is the numerator and which is the denominator? - correct answersdfbetween = numerator dfwithin = denominator What does the F-ratio tell you? - correct answersThe overall significance (are there differences between the treatments) Can you perform an ANOVA with unequal n numbers? - correct answersYes Why use an ANOVA rather than multiple t-tests to compare more than two groups? - correct answersYou must run multiple t-tests to compare more than two groups, and each time you run a test the chance of a type 1 error gets higher. ANOVA runs everything at the same time so the chance does not increase no matter how many groups you have. Are F-ratios and t-scores related? - correct answersYes, F = t² What are the three assumptions of an independent-measures ANOVA? - correct answers1. The observations within each sample must be independent. 2. The populations from which the samples are selected must be normal. 3. The populations from which the samples are selected must have equal variances (homogeneity of variance). Note: There are the same as the independent-samples t-test. What does T stand for in ANOVA? - correct answersT is the sum of X for each condition. What does G stand for in ANOVA? - correct answersG is the sum of all scores (sum of the T's) What does k stand for in ANOVA? - correct answersk is the number of levels Write out, in words, what the F-ratio numerator and denominator represent. Which do you want to be higher to find significance? - correct answersNumerator: Systematic differences caused by the treatments Denominator: Random, unsystematic differences Numerator should be larger to find significance. How does the ANOVA denominator change in a repeated-measures test? - correct answersThe denominator becomes the residual/error variance What does P stand for in ANOVA? When is it used? How do you find it? - correct answersP represents the total of all scores for each individual in the study and is used in repeated-measures ANOVAs. To find this, you just add up row totals. How do the ANOVA formulas change for repeated-measures and factorial ANOVAs? - correct answersRepeated-measures: The variable P is introduced and replaces T. Residual/error variance also becomes part of the denominator. Factorial: You now test for interactions (AxB) so there are more formulas that break down the between-subjects variance. What is a main effect? - correct answersThe mean differences among the levels of one factor (how much is A or B directly effecting the dependent variable) What is an interaction? - correct answersAn interaction between two factors occurs whenever the mean differences between individual treatment conditions are different from what would be predicted from the overall main effects of the factors. In other words, are A and B interacting and effecting one another? On a graph how do you tell if there is an interaction? - correct answersIf the lines are not parallel then there is an interaction because they will cross at some point. How many hypothesis tests are involved in a two-factor ANOVA? What are they? - correct answers1. The main effect for factor A. 2. The main effect for factor B. 3. The interaction (AxB). Are the assumptions for a two-factor ANOVA the same as a single-factor ANOVA? - correct answersYes. In a two-factor ANOVA, does the denominator or numerator change? - correct answersThe numerator, it uses numbers from Factor A, Factor B, or the AxB interaction Repeated Measures ANOVA - correct answersA one-way ANOVA that involves correlated groups of participants. independent measures anova - correct answers-uses multiple participant samples to test the treatments Levels (ANOVA) - correct answersthe individual conditions or values that make up a factor p-value alpha - correct answersreject the null hypothesis alpha level (level of significance) - correct answersa probability value that is us