Test Pilot 2.x EDPSY 507 Final Exam RLH EDPSY 507 Final Examination lAnswer the following questions. If your answer is not exact, choose the alternative closest to your answer. Regression is everything. @I hope you learned something in the class. Have a great summer. WHITE Helvetica,Arial 4 BLACK 7 1 ` ‚Use the following data to answer this question. Date = [v1], [v2], [v3], [v4], [v5] and the model is Bo = [v6] w
1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Congratulations ƒSorry your answer was incorrect. you simply need to subtract Bo from each score, square that difference and add them all together. ?ð @$ ?ð @" v1?ð @$ ?ð @" v2?ð @$ ?ð @" v3?ð @$ ?ð @" v4?ð @$ ?ð @" v5?ð @$ ?ð @" v6 ÿÿÿÿÿÿÿÿÿÿÿÿ?ð 9(v1-v6)^2 + (v2-v6)^2 + (v3-v6)^2 + (v4-v6)^2 + (v5-v6)^2 L([v1]-[v6])^2 + ([v2-[v6])^2 + ([v3]-[v6])^2 + ([v4]-[v6])^2 + ([v5]-[v6])^2 j €Using the following data, calculate the error where the model = the mean. Data = [v1], [v2], [v3], [v4], [v5] w
1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Congratulations hSorry you need to subtract the mean from every value, square that difference, and add them all together. ?ð @$ ?ð @" v1?ð @$ ?ð @" v2?ð @$ ?ð @" v3?ð @$ ?ð @" v4?ð @$ ?ð @" v5 ÿÿÿÿÿÿÿÿÿÿÿÿ?ð “(v1-((v1+v2+v3+v4+v5)/5))^2 + (v2-((v1+v2+v3+v4+v5)/5))^2 + (v3-((v1+v2+v3+v4+v5)/5))^2 + (v4-((v1+v2+v3+v4+v5)/5))^2 + (v5-((v1+v2+v3+v4+v5)/5))^2 Ï([v1]-(([v1]+[v2]+[v3]+[v4]+[v5])/5))^2 + ([v2]-(([v1]+[v2]+[v3]+[v4]+[v5])/5))^2 + ([v3]-(([v1]+[v2]+[v3]+[v4]+[v5])/5))^2 + ([v4]-(([v1]+[v2]+[v3]+[v4]+[v5])/5))^2 + ([v5]-(([v1]+[v2]+[v3]+[v4]+[v5])/5))^2 t vIf the compact model has error = [v1] and the augmented model has error = [v2], "What is the PRE =? w
1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Congratulations —Sorry, you subtract the error for the augmented model from the error for the compact model, and divide that result by the error from the compact model. @i @rÀ ?ð @Y v1@Y @hà ?ð @XÀ v2 ÿÿÿÿÿÿÿÿÿÿÿÿ@
(v1-v2)/v1 ([v1]-[v2])/[v1] ~ yUsing the following information calculate the F statistic. PRE = .52 PA = [PA], PC = [PC], and N = [N] w
1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Correct. uSorry you needed to take .52 and divide by the difference (PA - PC) then divide that result by .48 divided by N - PA. @ @ ?ð @ PA?ð @ ?ð ?ð PC?ð @I ?ð @H€ N ( ÿÿÿÿÿÿÿÿÿÿÿÿ?ð (.52/(PA-PC))/(.48/(N-PA)) "(.52/([PA]-[PC]))/(.48/([N]-[PA])) ˆ .Assume the following: Model C is L = bo + b1X1 + b2X2 and Model A is L = bo + b1X1 + b2X2 + b3X3 How many parameters are in the compact model? e1
2
3
4
1 2 3 4 0 0 1 0 6Sorry there are three parameters in the compact model. 6Sorry there are three parameters in the compact model. -Correct, the constant and the two predictors 6Sorry there are three parameters in the compact model. ,1 points awarded for selecting 3
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ ’ 0Assume the following: Model C is L = bo + b1X1 + b2X2 and Model A is L = bo + b1X1 + b2X2 + b3X3 How many parameters are in the augmented model? e1
2
3
4
1 2 3 4 0 0 0 1 :Sorry there are 4. The constant and the three predictors. :Sorry there are 4. The constant and the three predictors. :Sorry there are 4. The constant and the three predictors. Correct ,1 points awarded for selecting 4
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ œ AAssume the following: Model C is L = bo + b1X1 + b2X2 and Model A is L = bo + b1X1 + b2X2 + b3X3 How many degrees of freedom in the numerator of the F statistic? e1
2
3
4
1 2 3 4 1 0 0 0 Correct 4Sorry, there is only one degree of freedom (PA - PC) 4Sorry, there is only one degree of freedom (PA - PC) 4Sorry, there is only one degree of freedom (PA - PC) ,1 points awarded for selecting 1
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ ¦ ²Which of the following techniques tries to answer how many groups do I have represented in this data? ºDiscriminant Analysis
Hierarchical Cluster Analysis
K-means Cluster Analysis
Factor Analysis
Discriminant Analysis Hierarchical Cluster Analysis K-means Cluster Analysis Factor Analysis 0 1 0 0 ?Sorry, for this procedure you need to know the group structure. Correct GSorry, to use this procedure you need to know how many groups you want. 4Sorry, this procedure groups variables not subjects. H1 points awarded for selecting Hierarchical Cluster Analysis
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ ° °Which of the following techniques helps us determine how many things (concepts) are being measured? ºDiscriminant Analysis
Hierarchical Cluster Analysis
K-means Cluster Analysis
Factor Analysis
Discriminant Analysis Hierarchical Cluster Analysis K-means Cluster Analysis Factor Analysis 0 0 0 1 /Sorry, this procedure does not group variables. 9Sorry, this procedure groups subjects, and not variables. 9Sorry, this procedure groups subjects, and not variables. Correct :1 points awarded for selecting Factor Analysis
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ º ÇWhich of the following procedures allows you to answer the question of which subjects belong to a known number of groups? ºDiscriminant Analysis
Hierarchical Cluster Analysis
K-Means Cluster Analysis
Factor Analysis
Discriminant Analysis Hierarchical Cluster Analysis K-Means Cluster Analysis Factor Analysis 0 0 1 0 )Sorry, this is not the correct procedure. XSorry, this procedure determines the number of groups, which in this question was known. Correct XSorry, this procedure groups variables that share substantial variance with one another. C1 points awarded for selecting K-Means Cluster Analysis
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ Ä ¶In the discriminant analysis picture shown below, the line indicated by the red letter "A" is pointing to a centroid
a contour
a discriminant function line
None of these
@ ð discrim.jpg
a centroid a contour a discriminant function line
None of these 1 0 0 0 Correct Sorry ,A points to a centroid Sorry, A points to a centroid Sorry, A points to a centroid 51 points awarded for selecting a centroid
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ Î ÉLooking at the picture above illustrating a discriminant function analysis, the line marked by the red letter "B" points to ™a centroid
a contour
a discriminant function
none of these.
@ ð discrim.jpg
a centroid a contour a discriminant function none of these. 0 1 0 0 Sorry, this is a countour Correct Sorry, B points to a contour Sorry, B points to a contour. 41 points awarded for selecting a contour
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ Ø ¹Looking at the discriminant analysis picture shown above, the line indicated by the red letter "C" points to œa centroid.
a contour.
a discriminant function.
none of these.
@ ð discrim.jpg a centroid.
a contour. a discriminant function. none of these. 0 0 1 0 /Sorry, this is the discriminant function line. .Sorry, this is the discriminant function line. Correct .Sorry, this is the discriminant function line. C1 points awarded for selecting a discriminant function.
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ â ™Looking at the discriminant function figure shown above, one would conclude. fThe variance/covariance values for the variables are the same for both groups.
The variance/covariance values for the variables are different for both groups.
@ ð discrim.jpg NThe variance/covariance values for the variables are the same for both groups. OThe variance/covariance values for the variables are different for both groups. 1 0 BCorrect, the spread and direction of the group plots are the same. ‡Sorry, the spread and direction of the group plots are the same indicating that the variance/covariance matrices are the same for both. y1 points awarded for selecting The variance/covariance values for the variables are the same for both groups.
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ ì ÈFor a typical principal components factor analysis, the stopping rule is to stop factoring when the eigenvalue drops below: h.05
.1
1
2
.05 .1 1 2 0 0 1 0 #Sorry this is a typical alpha value Correct ,1 points awarded for selecting 1
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ ö æIn class, the thumb-rule value for determining whether a variable loading indicated that the variable should be included in naming a component factor was †below .05
.below 01
above ..05
above .40
below .05 .below 01
above ..05 above .40 0 0 0 1 Correct 41 points awarded for selecting above .40
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ ¤In conducting a 2X3 ANOVA, how many effect codes would the researcher need to generate? e6
5
4
3
6 5 4 3 0 1 0 0 ,1 points awarded for selecting 5
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ
-A researcher simply wants to test whether verbal IQ (VIQ) interacts with itself and improves the prediction of reading over a simple model where VIQ is used as the predictor. What is the compact model in this investigation? Read = bo + b1(VIQ) + b2(VIQ)(VIQ)
Read = Bo
Read = b1(VIQ)
Read = bo + b2(VIQ)
"Read = bo + b1(VIQ) + b2(VIQ)(VIQ) Read = Bo Read = b1(VIQ) Read = bo + b2(VIQ) 0 0 0 1 "Sorry, this is the augmented model ;Sorry, this is incorrect, there is no parameter estimated. +Sorry, there is no constant estimated here. Correct >1 points awarded for selecting Read = bo + b2(VIQ)
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ /A researcher simply wants to test whether verbal IQ (VIQ) interacts with itself and improves the prediction of reading over a simple model where VIQ is used as the predictor. What is the augmented model in this investigation? Read = bo + b1(VIQ) + b2(VIQ)(VIQ)
Read = Bo
Read = b1(VIQ)
Read = bo + b2(VIQ)
"Read = bo + b1(VIQ) + b2(VIQ)(VIQ) Read = Bo Read = b1(VIQ) Read = bo + b2(VIQ) 1 0 0 0 Correct M1 points awarded for selecting Read = bo + b1(VIQ) + b2(VIQ)(VIQ)
ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ 'If a researcher was using the following augmented model, and found it to be a significant improvement over the compact model, would the slope of the simple equation of X1 predicting Y be changing? Y = bo + b1X1 + b2X2 çYes
No
Yes No 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 GSorry, this is not an interactive model and the slope would not change. -1 points awarded for selecting No
ÿÿÿÿÿÿÿÿÿÿÿÿ 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Hale fanny1 h12@psu.edu