Help with 3 sections- 3 to 4 pages- statistical methods for criminal justice

Hypotheses

H1: There is an association between Personal Irresponsibility and Counselors Rapport.

H2: There is an association between Desire for Help and Counselors Rapport.

H3: There is an association between Treatment Readiness and Counselors Rapport.

H4: There is an association between Desire for Help and Personal Irresponsible.

H5: There is an association between Treatment Readiness and Desire for Help

H6: There is an association between Treatment Readiness and Personal Irresponsible

Data Management

In this study I computed a dependent variable Counselors Rapport and three independent variables Desire for Help, Personal Irresponsibility and Treatment Readiness.

I ran frequencies for variables CEST015, CEST021, CEST038, CEST042, CEST043, CEST050, CEST052, CEST063, CEST084, CEST110 and CEST128. There are three reasons why you run frequencies. The first reason is to see if all of the indicators are measured on the same scale. The second reason is to see if there are any responses outside of the valid or legitimate responses. The third reason is to see if everything is moving in the same direction. As a result of running frequencies and looking at our legitimate responses of all variables, we now know that the variables are all nominal. With all of the variables having the same value and label, we can tell that all of the variables are moving in the same direction.

Variables CEST015, CEST021, CEST038, CEST042, CEST043, CEST050, CEST052, CEST063, CEST084, CEST110 and CEST128 were factored into a factor analysis. A factor analysis is SPSS’s way of telling if the indicators are going to group well together, based on the responses. According to Extraction Sums of Squared Loadings under Total Variance Extraction, the number of total values listed tells if the indicators group well together. Because only one number appears under total, the factor analysis for these indicators has determined that the indicators are all grouping together under the same heading. Another sign of the indicators grouping well together is having no value for the rotated component matrix. The component matrix showed a negative value, so I did a reverse code with the variable CEST050 to CEST050r. Once the reverse code was done I ran a frequency to make sure the negative value wasn’t there anymore. The indicators are grouping well together and all fit into one idea of “RN”.

The Eigen Values under Component Matrix represents how good the actual indicators work with the rest. Any Eigen Value above .4 is good value for these indicators. All indicators have Eigen Values above .4 now. The recoded variable was run through a reliability analysis. Cronbach's Alpha tells how well the indicators group together. It gives an actual value to determine how got a fit they are. Typically a Cronbach's Alpha of .75 or higher is ideal. The Cronbach's Alpha for these indicators is .937, which is above the ideal value for the Cronbach’s Alpha.

The variable CR was created using CEST015, CEST021, CEST038, CEST042, CEST043, CEST052, CEST063, CEST084, CEST110, CEST128 and CEST050r.

I then ran frequencies CEST003, CEST032, CEST039, CEST065, CEST086, and CEST116. There are three reasons why you run frequencies. The first reason is to see if all of the indicators are measured on the same scale. The second reason is to see if there are any responses outside of the valid or legitimate responses. The third reason is to see if everything is moving in the same direction. As a result of running frequencies and looking at our legitimate responses of all variables, we now know that the variables are all nominal. With all of the variables having the same value and label, we can tell that all of the variables are moving in the same direction.

I didn’t have to recode any variables. All of these variables CEST003, CEST032, CEST039, CEST065, CEST086, and CEST116 were factored to create a factor analysis. A factor analysis is SPSS’s way of telling if the indicators are going to group well together, based on the responses. According to Extraction Sums of Squared Loadings under Total Variance Extraction, the number of total values listed tells if the indicators group well together. Because only one number appears under total, the factor analysis for these indicators has determined that the indicators are all grouping together under the same heading. Another sign of the indicators grouping well together is having no value for the rotated component matrix. These six chosen indicators all fit into one idea of “DH”.

The Eigen Values under Component Matrix represents how good the actual indicators work with the rest. Any Eigen Value above .4 is good value for these indicators. All indicators have Eigen Values above .4. The variables were ran through a reliability analysis. Cronbach's Alpha tells how well the indicators group together. It gives an actual value to determine how got a fit they are. Typically a Cronbach's Alpha of .75 or higher is ideal. The Cronbach's Alpha for these indicators is .715. This is above the ideal value for the Cronbach’s Alpha. The variable DH was created using six indicators: CEST003, CEST032, CEST039, CEST065, CEST086, and CEST116.

Frequencies were run for variables CEST024, CEST047, CEST055, CEST068, and CEST118. There are three reasons why you run frequencies. The first reason is to see if all of the indicators are measured on the same scale. The second reason is to see if there are any responses outside of the valid or legitimate responses. The third reason is to see if everything is moving in the same direction. As a result of running frequencies and looking at our legitimate responses of all variables, we now know that the variables are all nominal. With all of the variables having the same value and label, we can tell that all of the variables are moving in the same direction.

I didn’t have to recode any variables. All of these variables were factored into a factor analysis CEST024, CEST047, CEST055, CEST068, and CEST118. A factor analysis is SPSS’s way of telling if the indicators are going to group well together, based on the responses. According to Extraction Sums of Squared Loadings under Total Variance Extraction, the number of total values listed tells if the indicators group well together. Because only one number appears under total, the factor analysis for these indicators has determined that the indicators are all grouping together under the same heading. Another sign of the indicators grouping well together is having no value for the rotated component matrix. The five chosen indicators all fit into one idea of “TN”.

The Eigen Values under Component Matrix represents how good the actual indicators work with the rest. Any Eigen Value above .4 is good value for these indicators. The only indicator with an Eigen Value less than .4 is CEST118. In terms of frequency it says “need more medical care”.

The variables were run through a reliability analysis. Cronbach's Alpha tells how well the indicators group together. It gives an actual value to determine how got a fit they are. Typically a Cronbach's Alpha of .75 or higher is ideal. The Cronbach's Alpha for these indicators is .599. This is below the ideal value for the Cronbach’s Alpha. If CEST118 was removed the value of Cronbach’s Alpha would increase to .634. The decision was made to keep the indicator because more medical care is needed. The variable TN was created using 5 indicators: CEST024, CEST047, CEST055, CEST068, and CEST118.

The last set of frequencies I ran were CEST2003, CEST2004, CEST2029, CEST2039, CEST2045 and CEST2056. There are three reasons why you run frequencies. The first reason is to see if all of the indicators are measured on the same scale. The second reason is to see if there are any responses outside of the valid or legitimate responses. The third reason is to see if everything is moving in the same direction. As a result of running frequencies and looking at our legitimate responses of all variables, we now know that the variables are all nominal. With all of the variables having the same value and label, we can tell that all of the variables are moving in the same direction.

I didn’t have to recode any variables. All of these variables were factored into a factor analysis CEST2003, CEST2004, CEST2029, CEST2039, CEST2045 and CEST2056. A factor analysis is SPSS’s way of telling if the indicators are going to group well together, based on the responses. According to Extraction Sums of Squared Loadings under Total Variance Extraction, the number of total values listed tells if the indicators group well together. Because only one number appears under total, the factor analysis for these indicators has determined that the indicators are all grouping together under the same heading. Another sign of the indicators grouping well together is having no value for the rotated component matrix. The six chosen indicators all fit into one idea of “PI”.

The Eigen Values under Component Matrix represents how good the actual indicators work with the rest. Any Eigen Value above .4 is good value for these indicators. All the Eigen Value is above .4. The variables were run through a reliability analysis. Cronbach's Alpha tells how well the indicators group together. It gives an actual value to determine how got a fit they are. Typically a Cronbach's Alpha of .75 or higher is ideal. The Cronbach's Alpha for these indicators is .603. The variable PI was created using 6 indicators.

Data Analysis Comment by Davis-Ganao, Jessica S: More detail is needed

In the aspect of the Univariate descriptive analysis a SPSS frequency was run for the nominal and ordinal level measure. The measure of central tendency was used to describe the data by showing the mode, median and frequencies. A frequency was run to show how often something occurs. Also another Univariate descriptive analysis SPSS frequency was run for the interval and ratio level measure and I used the measure of dispersion to show the range of the minimum and maximum level of the variables along with mean and standard deviation.

For the Correlations Matrix I ran a bivariate descriptive Analysis to compare the independent variable to the independent variable or the independent variable to the dependent variable. This correlation matrix is run based off of an interval level measure.

I ran a multivariate analysis to show the t-test, anova, regression and partial correlation because they demonstrate 3 or more variables in the distribution. The t-test and the anova are interval/ ratio level measure because it reports the mean and standard deviation. A regression was also run and the components standardized beta and statistical significance are examined. A SPSS partial correlation was the last analysis run which pointed out the main effects of a coefficient correlation and the coefficient partial correlation control variable grade.

Findings

The tables below represent the findings of the Univariate, Bivariate and Multivariate Analysis in SPSS Output form. Comment by Davis-Ganao, Jessica S: More detailed transition

Univariate Findings

The first table represents the demographics of Counselor Rapport such as Job and Sex. The table uses nominal and ordinal level measures showed by the frequencies, percentages and measure of central tendency.

Table 1. Univariate Statistics for Demographics

As it relates to job, the median category is 2, which represents a part time job. Also there is a lot of fluctuation at the beginning and the end of the distribution. There are less part time workers than any job (n=185).As it relates to gender, males represent 56% (n=891) of the sample. The modal category is 1 which represents males. Although males are the modal category in terms of gender, females are almost equally represented in this data set.

The second table below displays the range, mean and standard deviation for the interval and ratio level measures in this distribution.

Table 2. Univariate Descriptive Findings for Interval and Ratio level Study Variables

a. The lower the number means that they don’t want counseling.

b. The lower the number the less they have a desire for help.

c. The lower the value the less treatment ready they are.

d. The higher the number the less personally irresponsible someone is.

e. The highest grade complete the lower the result.

As it relates to Counselor Rapport, the average counselor has a rapport of 49.15, a standard deviation of 10.578. The standard deviation shows that there is little fluctuation around the mean. The minimum score of the distribution is 13 and the maximum score is 65 which means the lower the number means that they don’t want counseling. As it relates to desire for help, the average is 28.83, with a standard deviation of 4.267. This means that there aren’t a lot of people who don’t desire help. The standard deviation shows that there is little fluctuation around the mean. The minimum score of the distribution is 6 and the maximum score is 30 which mean the lower the number the less they have a desire for help.

As it relates to the treatment readiness scale, the average is 16.59, with a standard deviation of 3.902. This means that the average respondent should average 16.59 on the scale. The standard deviation shows that there is little fluctuation around the mean. The minimum score of the distribution is 5 and the maximum score is 25 which means that on the scale the lower the number the less you are on the scale.

As it relates to personal irresponsibility scale, the average is 13.00, with a standard deviation of .936. This means that the average respondent scored a 13 on the personal irresponsibility scale. The standard deviation shows that there is little fluctuation around the mean. The minimum score is 6 and the maximum score is 30 which means the distribution on the scale go from high to low. As it relates to grade, the average respondent has a grade of 10.52, a standard deviation of 1.924. Which means the average person is a C student.

Bivariate Findings

This table is a correlation matrix that shows the main effects as well as the intercorrelations. The asterisks display the confidence level and statistical significance of each.

Table 3. Correlations Matrix for Counselor Rapport and Independent Variables

*p≤.05; **p≤=.01; ***p≤.001

The strongest relationship between the independent variable and dependent variable is that between Desire for help and Counselors Rapport (r=.338; p=.000). The more the respondent desired for help the more the counselor was able to build a rapport. The next strongest relationship personal irresponsibility and counselors rapport (r=-.253; p=.000).

The more personally irresponsible the respondent is the more it’ll help build a counselors relationship with the individual. The weakest relationship is treatment readiness and counselor rapport (r=.118; p=.000). Those who perceived to be treatment ready need more of a counselor’s relationship. All variables are significant. Overall the relationships between the independent variables are moderate to weak or show no relationship at all.

Multivariate Analysis Comment by Davis-Ganao, Jessica S: Transition needed here.

Partial Correlations

Test whether or not a control variable is an intervening, spurious, or no effect at all on the relationship between 2 variables.

Table 4. Partial Correlations with Counselors Rapports

*p≤.05; **p≤=.01; ***p≤.001

t-tests

This table shows t-test which, are used when you want to determine whether there is a difference between two groups on some given variables. In this table you will see the mean and standard deviation of males and females for each main effect.

Table 5. Multivariate results of T-tests with Study Variables

*p≤.05; **p≤=.01; ***p≤.001

In terms of the t-tests results, the findings indicate that there is a statistically significant difference between the means for sex on counselor rapport (t=-4.110; p=.000) for males (mean=48.16; SD=10.759) and females (mean=50.42; SD=10.221). There is a statistically significant difference between the means for sex on desire for help (t=-11.387; p=000) for males (mean=22.76; SD=4.377) and females (mean= 25.18; SD=3.707). There is a statistically significant difference for males and females on sex and treatment (t=-7.712; p=.000) males (mean=15.92; SD=3.900) and females (mean=17.45; SD=3.736). In addition, there is a statistically significant difference between sex and personal irresponsibility (t=5.156; p=.000) males (mean=13.46; SD=4.088) and females (mean=12.42; SD=3.681). Overall you can reject the null hypothesis for all of the variables.

ANOVA

This table discusses the results of an ANOVA; an ANOVA is run when you have 3-5 categories. In this table you will see the control variable Jobs broken down into three categories full time, part time and other by way of the mean and standard deviation and the F value.

Table 6. Results of the Anova with Study Variables

In terms of the ANOVA results, There is a statistically significant difference between the means for job on treatment readiness (F=8.714; p=.000). In contrast there is no statistically significant difference between the means for job on counselor rapport (F=.603; p=.547). There is no significant difference for jobs on desire for help (F=2.687; p= .068) or on personal irresponsibility (F=2.826; p=.060) Overall you can reject the null hypothesis for treatment readiness and accept the null hypothesis for counselor rapport, desire for help and personal irresponsibility.

Regression

In terms of the model for counselor rapport with variables desire for help, treatment readiness and personal irresponsibility the findings show a weak (r2=.127) and a significant relationship (f=67.361; p=.000). The strongest relationship in the model is that between desire for help and counselor rapport (std .beta=.283; p=.000). The next strongest relationship is that between personal irresponsibility and counselor rapport (std. beta=-.163; p= .000). The weakest relationship is between treatment readiness and counselor rapport (std. beta= -.042; p=.137), this relationship is not significant. Overall the findings show that there is support for the entire hypothesis except for treatment readiness and counselor rapport.

Table 7. Results of the OLS Regression Analysis with Counselor Rapport as the Dependent

Variable

*p≤.05; **p≤=.01; ***p≤.001

Discussion Comment by Davis-Ganao, Jessica S: This should be longer

All of the main effects projected in the hypothesis model are the same in the final model. The original model stated that there is an association between Desire for Help and Personal Irresponsible. The research states that, “Substance abuse affects entire families, yet only recently has attention been focused on the needs of children with parents who abuse drugs and/or alcohol (Greenburg, 2000).”

In previous research with another data set (Simpson et al., 1997e), we demonstrated the importance of treatment engagement- defined in terms of counseling session attendance and mutual ratings of the therapeutic relationship between counselor and client-for retention in a sample of methadone maintenance clients (Sindelar, 2001). The research collaborates with my hypothesis that there is an association between Treatment Readiness and Desire for Help.

There is an association between Treatment Readiness and Personal Irresponsible according to my research. Treatment reduces drug use and crime and increases individuals' functioning (Sindelar, 2001).

Model Revision

As predicted, desire for help, treatment readiness and personal irresponsibility all had an effect on counselor’s rapport. Reliable information stated that one’s desire for help will help build a relationship with a counselor. A counselor will help the respondent in need get help. The counseling relationship or therapeutic alliance is perceived to be central to achieving a positive outcome in all mental health counseling (Gelso & Fretz, 1992), and it is especially important that a positive relationship or therapeutic alliance be formed early in addictions counseling before the more difficult or challenging times (e.g., withdrawal symptoms, relapse) occur (Merta, 2001). The findings were moderately consistent with the original model. Based on my findings building a relationship with your counselor will help you if you’re desiring for help, ready for treatment and ready to be personally responsible for your actions. Because the counselor will help you transition to where you need to be.

Figure 2. Revised Model of Building Relationships

Personal Irresponsibility