$FL2@(#) IBM SPSS STATISTICS DATA FILE 64-bit MS Windows 20.0.0 ################ #########Y@20 Jun 1615:03:34 ###########################GROUPS ########################ABSENCES########################SAT ########################ACTUALMA############################################## ################ ############ ##############################################################;###GROUPS=Groups ABSENCES=Absences SAT=SAT ACTUALMA=ActualMark######################## ###################R###Groups:$@Role('0' )/Absences:$@Role('0' )/SAT:$@Role('0' )/ActualMark:$@Role('0' )############ ###windows1252#######fhef#####`@#####p@effg#####@@#####@@eeed#####@#####@f hfl#####p@#####~@fkeg##### |@#####@ $FL2@(#) IBM SPSS STATISTICS DATA FILE 64-bit MS Windows 20.0.0 ##########################Y@20 Jun 1615:02:16 ###########################DAY ####Day Reading ########################TEMPERAT####Average Temperature ########################CONSUMPT####Average Water Consumption ############################################################## ############ ########################################### #######1###DAY=Day TEMPERAT=Temperature CONSUMPT=Consumption############################################B###Day: $@Role('0' )/Temperature:$@Role('0' )/Consumption:$@Role('0' )############ ###windows-1252#######e|tfxg}hijk### University of Calgary in Qatar Nursing 609: Applied Statistics for Nursing Research Data Analysis 4.1: Linear Regression Analysis 1. Compute a simple linear regression with one independent variable Open Datafile - waterconsumptionregression.sav in D2L Day Temperature 1 2 3 4 5 6 7 24 28 29 29 33 36 37 45 Actual Water Consumption 16 20 25 27 32 48 48 Predicted Water Consumption H1: Temperature is a good and significant predictor of water consumption. Run a simple linear regression analysis in SPSS using temperature (independent variable) to predict water consumption (dependent variable) Analyzeregressionlinear Dependent variable: water consumption Independent Variables: temperature a) Report the Coefficients table below. Make sure to add R2. Write one or two sentences explaining this result, and include the F ratio and p value from the ANOVA table. What is the regression formula for this model? Report it below. Y=a+bxX Y = Constant + beta1 (X) Y = ________ + _______X <-- So, if you know X, you can estimate Y b) Estimate water consumption for a day that is 37 degrees. _____________ Is it the same as the actual temperature recorded for 37 degress in our data? No. Remember, it is an estimate. However, it should be relatively close. c) Estimate water consumption for a day that is 45 degrees. _____________ 2. Compute a multiple linear regression with 2 independent variables Open Datafile - linearregressionGroupsAbsencesSATMarks.sav H1: Number of Absences and SAT are a good and significant model for predicting actual grades a) Run a multiple regression analysis (Analyze --> Regression --> Linear; Choose Mark as DV and SAT and Absences as IVs) Report the results below. Make sure to include the Coefficients table, at R2, and report the F-ratio and its p-value in the explanation. b) report the regression formula below: Y = Constant +beta1 (X1) + beta2 (X2) Y= ____________ + ____________ + _______________ c) Use the above formula to estimate marks for the following studentsI i) 6 absences and a 420 SAT score ii) 0 absences and a 510 SAT score iii) 10 absences and a 780 SAT score 3. Compute a Stepwise Linear Regression This tests 2 models: 1. SAT and 2. SAT, absences to predict mark Analyzeregressionlinear Dependent variable: Actual Grade Independent Variables: SAT, Absences Select Method = STEPWISE. Report the best model below. Include the coefficients table, with R2 added. Explain the results, including the F ratio and p-value. $FL2@(#) IBM SPSS STATISTICS DATA FILE 64-bit MS Windows 20.0.0 ##########################Y@20 Jun 1615:04:23 ###########################IQUCQ ############################################################## ################################ ###########IQUCQ=IQUCQ################################################IQUCQ: $@Role('0' )############ ###windows-1252######### $FL2@(#) IBM SPSS STATISTICS DATA FILE 64-bit MS Windows 20.0.0 ##########################Y@20 Jun 1615:06:40 ###########################SCORE ####Applied statistics test scores ########################CLASS ###Class Number############################################################## ############################################ ###########SCORE=score CLASS=class############################################ %###score:$@Role('0' )/class:$@Role('0' )############ ###windows1252#######neoepenereqepemelekewfxfxfufvfwftfufwfvfsgrgugtgpgpgsgrgtgvg#### University of Calgary in Qatar Nursing 609: Applied Statistics for Nursing Research Data Analysis 4.2: Linear Regression Analysis 1. Use a 1-sample t-test to compare the mean for a small sample of UCQ students' IQ with the known mean for the population. Open Datafile - 1samplettestIQ.sav in D2L IQUCQ 105 106 102 103 110 108 a) Write an appropriate HA for this study: SPSS Steps: Open 5.2 1samplettestIQ.sav Compare the mean for the UCQ students in this data file with the known mean for the population (100). Analyze compare means one sample t-test enter 100 as the test value - this is the mean of the population ( ) we will compare with x b) Copy/paste the One Sample Test table below. c) Interpret and explain the results. Remember to include t and p values in your explanation (t=___, p=___). 2. Compare means: t-test Researchers want to test the effectiveness of method to reduce stress in pre-school children about to undergo the finger-stick procedure for a hematocrit (Hct) determination. Open Datafile - independentttestHematocrit.sav NOTE: Hct is the volume of blood sample occupied by cells. a) HA: Analyze Compare Means independent samples t-test Place BP Prior in the Test Variable box Click 'Define Groups' Type 1 for Group 1 (this is the treatment group) and type 2 for Group 2 (our control group) Click OK b) Copy/paste the Independent Samples Test table below. c) Check the Levene's Test for Equality of Variance. Report the F and p value below. d) Interpret and report the results of the t-test. Remember to draw conclusions and explain regarding the hypotheses. Make sure you use the correct t and p values, depending on the Levene's test result. Also, remember to include the t and p values in the explanation, as well as the group means. 3. Paired Sample t-test Researchers want to know if a semester of English classes has improved students' listening abilities. They give students a listening test at the beginning of the semester and then repeat it at the end. a) HA: SPSS Steps: Analyze Compare Means Paired-sample t-test Variable 1: pre-test Variable 2: post-test Click OK b) Copy/paste the Paired Samples Test below. c) Interpret and explain the results. Make sure to include the means, t and p values. Remember to draw a conclusion regarding the hypotheses. 4. Paired Sample t-test Instructors want to test for a significant difference in final exam scores in an Applied Statistics course. There were three different sections of the course taught that semester. a) HA: SPSS Steps: Analyze Compare Means One-way ANOVA Dependent: Applied Statistics test scores - this tells SPSS our dependent variable is test scores Factor: Class - this tells SPSS we are testing for a difference between these groups (the 3 different STAT 205 sections) Click OK b) Copy/paste the ANOVA table below. c) Interpret and explain the results of the ANOVA. Remember to include F and p. Also, remember to draw a conclusion regarding the hypotheses. NOTE: Can we tell which class results were significantly different from one another? For that, we need to conduct post hoc (or follow-up) t-tests. NOTE: In the future, you can set these up before doing the ANOVA the first time, and get all the results at once. SPSS Steps: Analyze Compare Means One-way ANOVA Make sure Dependent and Factor are set up the same as before (stat test scores and class, respectively) Click Post Hoc and select Scheffe Set the Significance Level to 0.05 (or whatever you want alpha to be) Click OK d) Copy/paste the Multiple Comparisons table below. e) Interpret/explain the results. Make sure to mention the type of post-hoc test done and the alpha value. You only need to mention the group differences that were significant. If none are significant, you don't have to report anything (ANOVA results would have been above .05 and we would have accepted the null hypothesis). $FL2@(#) IBM SPSS STATISTICS DATA FILE 64-bit MS Windows 20.0.0 ##########################Y@20 Jun 1615:05:01 ###########################BP %###BP Prior to hematocrit determination ########################GROUP ####Group Assignment##############? Experimental #######@#Control############################################################## ############ ####################################### ################BP=BP GROUP=Group############################################"###BP:$@Role('0' )/Group:$@Role('0' )############ ###windows-1252#######eeeeeeeeeeffffffffff $FL2@(#) IBM SPSS STATISTICS DATA FILE 64-bit MS Windows 20.0.0 ################ #########Y@20 Jun 1615:05:58 ###########################PRETEST ####Pretest beginning of term ########################POSTTEST####Post test end of term ############################################################## ###################################################!###PRETEST=Pretest POSTTEST=posttest######################## ###################*###Pretest:$@Role('0' )/posttest:$@Role('0' )############ ###windows-1252#######nwoxpxnurvqwptmulwkv#### NURS 609 Data Analysis Assignment 4 4.1 1. Compute a simple linear regression with one independent variable Open Datafile - waterconsumptionregression.sav in D2L Day Temperature 1 2 3 4 5 6 7 24 28 29 29 33 36 37 45 Actual Water Consumption 16 20 25 27 32 48 48 Predicted Water Consumption H1: Temperature is a good and significant predictor of water consumption. Run a simple linear regression analysis in SPSS using temperature (independent variable) to predict water consumption (dependent variable) Analyzeregressionlinear Dependent variable: water consumption Independent Variables: temperature a) Report the Coefficients table below. Make sure to add R2. Write one or two sentences explaining this result, and include the F ratio and p value from the ANOVA table. Coefficientsa Model Unstandardized Coefficients Standardized t Sig. Coefficients B 1 (Constant) Average Temperature Std. Error -50.934 9.234 2.651 .296 Beta .970 -5.516 .003 8.945 .000 a. Dependent Variable: Average Water Consumption b. R2 = 0.941 Average temperature is a good and significant prediction of water consumption (F= 80, 01, P= 0.00) Temperature would seem to explain 94% of variance in water consumption. What is the regression formula for this model? Report it below. Y=a+bxX Y = Constant + beta1 (X) Y = -50.934+ 2.651X <-- So, if you know X, you can estimate Y b) Estimate water consumption for a day that is 37 degrees. -50.934+ 2.651(37) = 47.153 _____________ Is it the same as the actual temperature recorded for 37 degress in our data? No. Remember, it is an estimate. However, it should be relatively close. c) Estimate water consumption for a day that is 45 degrees. -50.934+ 2.651(45) = 68.361 _____________ 2. Compute a multiple linear regression with 2 independent variables Open Datafile - linearregressionGroupsAbsencesSATMarks.sav H1: Number of Absences and SAT are a good and significant model for predicting actual grades a) Run a multiple regression analysis (Analyze --> Regression --> Linear; Choose Mark as DV and SAT and Absences as IVs) Report the results below. Make sure to include the Coefficients table, at R2, and report the F-ratio and its p-value in the explanation. Coefficientsa Model Unstandardized Coefficients Standardized t Sig. Coefficients B (Constant) 1 SAT Absences Std. Error 33.422 13.584 .094 .021 -3.340 .773 a. Dependent Variable: ActualMark b. R2= .931 Beta 2.460 .043 .558 4.569 .003 -.527 -4.320 .003 Number of Absences and SAT are a good and significant model for predicting Actual grades (F=47,067, p=.00). The two independent variables determined 93% of variance in actual grades. b) report the regression formula below: Y = Constant +beta1 (X1) + beta2 (X2) Y= 33.422 + .094(X1) + -3.340(X2) c) Use the above formula to estimate marks for the following studentsI i) 6 absences and a 420 SAT score Y= 33.422 + .094(X1) + -3.340(X2) 33.422 + .094(420) + -3.340(6) Y= 33.422 +39.48-20.04 Y= 52.862 ii) 0 absences and a 510 SAT score Y= 33.422 + .094(X1) + -3.340(X2) 33.422 + .094(510) + -3.340(0) Y= 33.422+47.94-0 Y= 81.362 iii) 10 absences and a 780 SAT score Y= 33.422 + .094(X1) + -3.340(X2) 33.422 + .094(780) + -3.340(10) Y= 33.422+73.32-33.4 Y= 73.342 3. Compute a Stepwise Linear Regression This tests 2 models: 1. SAT and 2. SAT, absences to predict mark Report the best model below. Include the coefficients table, with R2 added. Explain the results, including the F ratio and p-value. Model Summary Model R 1 2 R Square Adjusted R Std. Error of the Square Estimate .864 a .746 .715 8.470 .965 b .931 .911 4.729 a. Predictors: (Constant), SAT b. Predictors: (Constant), SAT, Absences C. R2= .931 Coefficientsa Model Unstandardized Coefficients Standardized t Sig. Coefficients B 1 (Constant) SAT (Constant) 2 SAT Absences Std. Error -7.526 17.426 .146 .030 33.422 13.584 .094 .021 -3.340 .773 Beta -.432 .677 4.851 .001 2.460 .043 .558 4.569 .003 -.527 -4.320 .003 .864 a. Dependent Variable: ActualMark R2= .931 SAT and number of absences are a good and significant predictor of students actual marks (F (2, 7) =47.07, p=.00). These two independent variable explain approximately 93% of students actual marks (R2=.931). Data Analysis 4.2 1. Use a 1-sample t-test to compare the mean for a small sample of UCQ students' IQ with the known mean for the population. Open Datafile - 1samplettestIQ.sav in D2L IQUCQ 105 106 102 103 110 108 a) Write an appropriate HA for this study: SPSS Steps: Open 5.2 1samplettestIQ.sav Compare the mean for the UCQ students in this data file with the known mean for the population (100). Analyze compare means one sample t-test enter 100 as the test value - this is the mean of the population ( ) we will compare with x One-Sample Statistics N IQUCQ Mean 6 Std. Deviation 105.67 Std. Error Mean 3.011 1.229 One-Sample Test Test Value = 100 t df Sig. (2-tailed) Mean Difference 95% Confidence Interval of the Difference Lower IQUCQ 4.610 5 .006 5.667 Upper 2.51 8.83 A) Ha: UCQ student IQ is significant higher than general population mean. B) We accept the H0, t=4.610, p= .006 C) UCQ student IQ(x-bar=105.67) a significantly higher than the population mean(m=100) (t= 4.61,p=.01) b) Copy/paste the One Sample Test table below. c) Interpret and explain the results. Remember to include t and p values in your explanation (t=___, p=___). 2. Compare means: t-test Researchers want to test the effectiveness of method to reduce stress in pre-school children about to undergo the finger-stick procedure for a hematocrit (Hct) determination. NOTE: Hct is the volume of blood sample occupied by cells. a) HA: The treatment group has significantly lower BP than control group. b) Copy/paste the Independent Samples Test table below. Independent Samples Test Levene's Test t-test for Equality of Means for Equality of Variances F Sig. t df Sig. Mean Dif- Std. Error 95% Confidence (2- ference Difference Interval of the Dif- tailed) ference Lower Equal variBP Prior to hematocrit determination ances assumed Equal variances not assumed .134 .719 1.847 1.847 Upper 18 .081 -10.000 5.414 -21.374 1.374 17.948 .081 -10.000 5.414 -21.377 1.377 c) Check the Levene's Test for Equality of Variance. Report the F and p value below. For the Levene's Test, equal variance is assumed, (F=.134 and p=.719) d) Interpret and report the results of the t-test. Remember to draw conclusions and explain regarding the hypotheses. Make sure you use the correct t and p values, depending on the Levene's test result. Also, remember to include the t and p values in the explanation, as well as the group (means=95) and control group(means=105)(t=-1.85,p=.08) Independent t test resulted indicate there is no significant difference between BP of the treatment group (mean =95) and control group (mean=105) (t=-1.847, p=.081) 3. Paired Sample t-test Researchers want to know if a semester of English classes has improved students' listening abilities. They give students a listening test at the beginning of the semester and then repeat it at the end. a) HA: The result of listening of post-test significant higher than the result of listening of pre-test. b) Copy/paste the Paired Samples Test below. Paired Samples Test Paired Differences t df Sig. (2tailed) Mean Std. Devi- Std. Error 95% Confidence Inter- ation Mean val of the Difference Lower Pair 1 Upper Pretest beginning of term - Post test -7.700 2.497 .790 -9.486 -5.914 -9.753 9 .000 end of term c) Interpret and explain the results. Make sure to include the means, t and p values. Remember to draw a conclusion regarding the hypotheses. Paired Samples t-test indicate the post-test listening (mean=18.3) scores were significantly higher than the pre-test (mean=10.6) (t=-9.753, p=.000) 4. 1-Way ANOVA Instructors want to test for a significant difference in final exam scores in an Applied Statistics course. There were three different sections of the course taught that semester. a) HA: There will be a significant different between final exam score between three groups. b) Copy/paste the ANOVA table below. ANOVA Applied statistics test scores Sum of Squares Between Groups Within Groups Total df Mean Square 297.800 2 148.900 95.400 27 3.533 393.200 29 F Sig. 42.142 .000 c) Interpret and explain the results of the ANOVA. Remember to include F and p. Also, remember to draw a conclusion regarding the hypotheses. Result of the one-way ANOVA suggest there is a significant difference between 2 or more class sections in final test scores (F (2, 27) =42.14, p=.00) NOTE: Can we tell which class results were significantly different from one another? For that, we need to d) Copy/paste the Multiple Comparisons table below. Multiple Comparisons Dependent Variable: Applied statistics test scores Scheffe (I) Class Number (J) Class Number Mean Difference Std. Error Sig. (I-J) 1 2 3 95% Confidence Interval Lower Bound Upper Bound 2 -7.700* .841 .000 -9.88 -5.52 3 -4.300* .841 .000 -6.48 -2.12 1 7.700 * .841 .000 5.52 9.88 3.400 * .841 .002 1.22 5.58 1 4.300 * .841 .000 2.12 6.48 2 -3.400* .841 .002 -5.58 -1.22 3 *. The mean difference is significant at the 0.05 level. e) Interpret/explain the results. Make sure to mention the type of post-hoc test done and the alpha value. You only need to mention the group differences that were significant. If none are significant, you don't have to report anything (ANOVA results would have been above .05 and we would have accepted the null hypothesis). Post-hoc analysis found that all three groups' test score were significantly different (Scheffe, p<.05). The mean of class 2 is the highest, while of the mean class1 is the lowest. Class 2 had a significantly higher test score than class3. Class 3 had a significantly higher score than class1