SPSS Assignment 4 WORD DOC

doc

School

University of Maryland, College Park *

*We aren’t endorsed by this school

Course

300

Subject

Statistics

Date

Jan 9, 2024

Type

doc

Pages

Uploaded by JudgePelicanPerson377

CCJS 300 Fall 2023 SPSS Assignment 4 (Due December 6 th , 2023 by 5pm) Your fourth SPSS assignment involves generating frequency distributions, inferential statistics, hypothesis tests, and writing theoretical essays on predictor variables. Name: Please read the following honor code and sign below: I pledge on my honor that I have not given nor received any unauthorized assistance on this assignment. Signed: Instructions: Overview: You will be generating frequency distributions and inferential statistics for 6 variables (5 independent variables and 1 dependent variable) found in the “Monitoring the Future 2007.sav” data file on the class Canvas page under “SPSS Assignment 4”. Then, you will carefully consider which 3 other variables you think would also work as independent (predictor) variables and write an essay for each concerning why you think those particular variables would be good predictors. You may want to review your notes and the textbook section concerning independent and dependent variables (in Hagan, chapter 1). The dependent variable is the variable that is influenced (predicted) by the independent variables. Dependent Variable: The dependent variable that you will use in this assignment is named V115 ; its variable label is “072B07A:#XMJ+HS/LIFETIME”. The information for this variable is on page (46) 50 of the codebook. [The number in parentheses is the number seen in the codebook , the other number is the number of the page when the codebook is opened as a .pdf file]. Independent Variables: The independent variables that you will use are named and labeled as follows, the codebook pages are also listed. V102 , “072B02 : :#CIGS SMKD/30DAY”(33) 37 V103 , “072B03:EVER DRINK” (34) 38 V150 , “072C03 :R'S SEX” (79) 83 V169 , “072C13B:R'ATTND REL SVC” (91) 95 V194 , “072C25 :#X/AV WK GO OUT” (116) 120 1

Statistical task: Frequency Distributions for all 6 variables : Please select the following statistics to be displayed with your frequency distributions from the Statistics menu: Mean, Median, Mode, and Range. Inferential Statistical Procedure: The inferential procedure you will use for this assignment is called regression . It is a procedure that allows you to select a dependent variable to be predicted by an independent variable. Technically you are performing multiple regression , since you are using more than one independent variable . For each independent variable included in the procedure various statistics are generated—more about these later when I discuss the output. Statistical task: Regression Analyze…Regression…Linear… Put V115 in the ‘Dependent:’ box. Put V102, V103, V150, V169, and V194 in the ‘Independent(s)’ box. Then click OK. Understanding and Interpreting the Regression Output You have seen plenty of frequencies output; we will focus on the regression output. In the table labeled ‘Model Summary’ note that the ‘R’ value is .598, and the ‘R Square’ value is .358. It’s true; .598 2 = .358. I’ll get back to this later. In the table labeled ‘ANOVA’ we can see that ‘F’ statistic is large and that the ‘Sig.’ Equals .000. This means that the overall regression model shows statistical significance. So far so good. In the final table labeled ‘Coefficients’ notice the three right side columns labeled ‘Standardized Coefficients: Beta’, ‘t’, and ‘Sig.’. Except for the top number in the ‘Sig.’ Column (which you can ignore), all of the other numbers are ‘.000’. This means that all of the variables in the regression procedure have attained statistical significance—all of the independent variables are ‘statistically significant’ predictors of a respondent’s number of lifetime uses of marijuana or hashish. The column labeled ‘t’ is the good old t statistic, the test associated with those significant variables. The column labeled ‘Standardized Coefficients: Beta’ shows the relative importance and direction (positive or negative relationship) of the predictor variables. We know they are all statistically significant predictors; this number gives us even more information. Notice that the most important variable (.402) is how much the respondent has smoked cigarettes in the last 30 days. The next most important (.231) variable is whether the respondent has ever had alcohol to drink. The least important variable (.054) is the gender of the respondent. The last two variables show an interesting pattern. They have almost the same importance (-.135 and . 129), but function in opposite ways. The more a respondent attends religious services the fewer number of occasions they claim to have used marijuana and hashish in their lifetime, this is a negative relationship. The more a respondent goes out in the evenings for ‘fun and recreation’, the more marijuana and hashish they claim to have smoked, this is a positive relationship. 2

Please examine the first ‘Coefficients’ table from your output carefully and copy the information into the table provided below. Variable Standardized Coefficient Beta t statistic value Sig. 072B02 : :#CIGS SMKD/30DAY 0.402 45.695 <0.001 072B03:EVER DRINK 0.231 26.113 <0.001 072C03 :R'S SEX 0.054 6.325 <0.001 072C13B:R'ATTND REL SVC -0.135 -15.707 <0.001 072C25 :#X/AV WK GO OUT 0.129 14.798 <0.001 Understanding the R Square In the ‘Model Summary’ table, we noted that the R value is .598, and the R Square value is .358. The R Square is the more important number; it represents the proportion of variance in the dependent variable (the number of occasions the respondent reported using marijuana or hashish in their lifetime) that we can account for and thus understand when using this set of predictor variables. We have a good regression model—all of the variables are significant predictors, but we have only accounted for 35.8% of the variance in the dependent variable! The other 64.2% of the variance remains unexplained, shrouded in mystery. Your next task for this assignment is to go to the codebook and find 3 more variables that we did not use as predictors and explain why you think they would be good, statistically significant predictors, predictors that would increase the R Square, the amount of variance we understand within the model. Please note these variables and why you think they would be good predictors on the remaining pages of the assignment. Now that you have chosen your 3 variables, let’s see if whether or not they actually are statistically significant predictors and how their addition changes the impact of the other variables. Statistical task: Regression Analyze…Regression…Linear… Put V115 in the ‘Dependent:’ box. Put V102, V103, V150, V169, V194 and your chosen 3 variables in the ‘Independent(s)’ box. Then click OK. Now examine the second ‘Coefficients’ table carefully and copy the information into the table provided at the top of the next page. 3

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Variable Standardized Coefficient Beta t statistic value Sig. 072B02 : :#CIGS SMKD/30DAY 0.374 41.654 <0.001 072B03:EVER DRINK 0.228 25.850 <0.001 072C03 :R'S SEX 0.053 6.234 <0.001 072C13B:R'ATTND REL SVC -0.123 -14.137 <0.001 072C25 :#X/AV WK GO OUT 0.130 14.958 <0.001 072B08A:#X LAS/LIFETIME 0.107 12.371 <0.001 072C07Cb(R):R’SHSHLD FATHER -0.044 -5.121 <0.001 072C12 :R’POL BLF RADCL 0.027 3.172 0.002 Important Note: For your variables to be considered significant predictors, the ‘Sig.’ values do NOT have to be .000 as they were in the original equation. As long as they are . 05 or, less your variables will be considered significant predictors. Your last task is to write an essay on whether or not your variables were ‘statistically significant’ predictors and how their inclusion into the model changed the total R Square value and the standardized beta coefficients of the other variables. Checklist for a completed assignment 1) Frequency distributions for all original 6 variables, including the statistics Mean, Median, Mode, and Range. 2) Regression procedure for original 6 variables (as shown in class). 3) First ‘Coefficients’ Table completely filled out. 4) 3 additional variables selected, named, and explained. 5) Regression procedure for 9 variables 6) Second ‘Coefficients’ Table completely filled out. 7) Essay (last page) on your variables and how they worked as predictors. 4

1) Which variable did you choose? Please list the variable name or label below. Variable name and label : V118 LSD USE LIFETIME Why do you think this variable would be a statistically significant predictor of the number of occasions a respondent has used marijuana or hashish in their lifetime? LSD is psychedelic drug and marijuana has been linked to have psychedelic effects but not complete considered a psychedelic drug. It is fair to say that majority of LSD users have used or still use marijuana in their life but, not all those who use marijuana also use LSD. I thought it would be a good predictor variable because of this connection to LSD users using marijuana being a large percentage of users. Both of these drugs share risk factors in individuals such as their environment and social circles since marijuana and LSD are similar in effects, and would attract similar people who are looking for the said effects. I expect LSD to have a strong positive correlation to number of occasions a respondent has used marijuana or hashish in their lifetime. __________________________________________________________________ 2) Which variable did you choose? Please list the variable name or label below. Variable name and label : V155 FATHER IN HOUSEHOLD Why do you think this variable would be a statistically significant predictor of the number of occasions a respondent has used marijuana or hashish in their lifetime? I would assume that a father in the household would have a strong negative correlation to number of occasions a respondent has used marijuana or hashish in their lifetime. A father in a household provides stability, which has been linked to deter a child’s engagement in risky behaviors. Having a father in the household also contributes to the monitoring and supervision of a child, which can limit the child’s access to marijuana and social circles, where marijuana or hashish is present. All of these factors of 5

having a father in the household will greatly reduce the number of times a respondent has ever used marijuana or hashish. __________________________________________________________________ 3) Which variable did you choose? Please list the variable name or label below. Variable name and label : V167 POLITICAL BELIEFS Why do you think this variable would be a statistically significant predictor of the number of occasions a respondent has used marijuana or hashish in their lifetime? I think political beliefs would have a strong positive correlation with the number of occasions a respondent has used marijuana or hashish in their lifetime. We have seen in recent years that marijuana has become legalized in 40 out of 50 states. With this new policy decisions, we can assume to be a shift in cultural attitudes regarding marijuana use. This can lead respondents to be more truthful in their answers giving use an accurate number. Political beliefs can also predict that majority of voters like marijuana and potentially use it, we know this because of the fact it has been able to be voted for legalization. __________________________________________________________________ 4) Overall, were your variables statistically significant predictors? How did their inclusion into the model change the R Squared value? Did the other beta coefficients change as well—describe this. All of my variables were statistically significant predictors in the number of occasions a respondent has used marijuana or hashish in their lifetime. The inclusion of my three variables raised the R squared value from 0.358 to 0.369. This is good because it allows 6

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

us to further understand the variables accounted for. Before my three variables there was an unexplained 64.2% of the variance and I lowered that to 63.1% of unexplained variance. The most important beta coefficient remains as if the respondent smoked cigarettes in the past 30 days followed by if the respondent has ever drunk alcohol, which is similar to the first linear regression test. The beta coefficient for LSD lifetime use was 0.107 which is moderate between the other variables. The beta coefficient for father in household was -0.044 which a pretty weak negative correlation. Before I ran my chosen variables the respondent’s sex was the least important variable in prediction, but now political beliefs have become the least important, with a beta coefficient of 0.027. Overall, the important variables found in this regression study are cigarettes smoked in the past 30 days, ever drunk alcohol, going out on the weekend, religious attendance, and LSD lifetime use. The least important variables found on this regression study was political beliefs, father in the household, and the respondents sex. This assignment is worth 12% of your total grade. 7

SPSS Assignment 4 WORD DOC

Related Documents