Week 6 Assignment
A Survey of 50 Clients
Fifty clients of LIGHT ON ANXIETY were surveyed regarding their satisfaction with services. The clients filled out the survey on completion of treatment in January. In June, the clients were telephoned and re-surveyed and were asked to rate their overall satisfaction again.
Variables in the Working File
Variable
Position
Label
Measurement Level
Description
Participantid
1
ID
Scale
Participant ID number
Intake
2
Intake experience
Scale
On a scale of 1 to 10, how would you rate the intake
experience?
Indcouns
3
Individual Counseling
Scale
On a scale of 1 to 10, how would you rate your
satisfaction with the individual counseling sessions?
Groupcouns
4
Group Counseling
Scale
On a scale of 1 to 10, how would you rate your
satisfaction with the group counseling sessions?
Pricefair
5
Fairness of sliding scale
Scale
On a scale of 1 to 10, how would you rate your
satisfaction with the sliding scale method of payment?
NewPatient
6
Type of Patient
Ordinal
0 = first time 1 = repeat admission
Usage
7
Usage Level
Scale
What percent of your mental health services are
provided by this center?
Satjan
8
Overall Satisfaction in January
Scale
On a scale of 1 to 7, rate your overall satisfaction
with your MHMR experience.
Satjun
9
Overall Satisfaction in June
Scale
On a scale of 1 to 7, rate your overall satisfaction
with your MHMR experience.
Court
10
Court ordered treatment
Nominal
Was your treatment court-ordered?
0 = No; 1 = Yes
Therapytype
11
Individual or family therapy
Nominal
0 = Individual; 1 Family
Preexist
12
Pre-existing Condition
Nominal
1 = Mental health; 2 = Substance Abuse; 3 = Both
INSTRUCTIONS:
For each research question, describe in your word document the application of the seven steps of the hypothesis testing model.
Step 1: State the hypothesis (null and alternate)
Step 2: State your alpha (unless requested otherwise, this is always set to alpha = .05)
Step 3: Collect the data (use one of the data sets).
Step 4: Calculate your statistic and p value (this is where you run SPSS and examine your output files).
Step 5: Retain or reject the null hypothesis. (This is where you report the results of your analyses t (df) = t value, p = sig. level).
Step 6: Assess the Risk of Type I and Type II Error (did the data meet the assumptions of the statistic; effect size; and sample size).
Step 7: State your results in APA style and format. Be sure to report whether any assumptions were violated. Also report post-hoc test findings when the overall ANOVA is significant. Be sure to also include relevant figures.
Research Questions
Question 1: Are there differences in satisfaction with the intake process of clients who admit with pre-existing mental health problems, substance abuse problems, or both?
1. Run the One-Way ANOVA. Click on ANALYZE/COMPARE MEANS/ONE-WAY ANOVA
2. Use Preexisting condition (Preexist) as the independent variable.
3. Use Usage Level (Usage) as the dependent variable.
4. Select descriptive statistics. Under Options, check the boxes for homogeneity of variance test and Welch.
5. We can also get a graph of the means of our groups, if we click on OPTIONS and then MEANS PLOT in the next dialog box (note: it is interesting to see how SPSS will automatically generate the y-axis range according to the data, this feature can make a nonsignificant result look significant and a significant result look nonsignificant depending on your data).
6. Generate post-hoc comparison to evaluate the differences between groups. Click on Post-hoc and check the box next to Tukey.
Step 1: State the hypothesis (null and alternate)
Ø Null Hypothesis (H0): Clients with a history of mental health issues, drug addiction issues, or both reports no discernible differences in their level of satisfaction with the intake procedure.
Ø Alternate Hypothesis (H1): Clients with a history of mental health issues, drug addiction issues, or both reports significantly different levels of satisfaction with the intake procedure.
Step 2: State your alpha (unless requested otherwise, this is always set to alpha = .05)
Ø Alpha (α): 0.05
Step 3: Collect the data (use one of the data sets).
Ø Use the provided data set with variables Preexisting condition (Preexist) as the independent variable and Usage Level (Usage) as the dependent variable.
Step 4: Calculate your statistic and p-value (this is where you run SPSS and examine your output files).
Oneway
[DataSet1] D:RSM701LOA3.sav
Descriptives
Usage Level
N
Mean
Std. Deviation
Std. Error
95% Confidence Interval for
Mean
Minimum
Maximum
Lower Bound
Upper Bound
Mental Health
18
35.833
5.1478
1.2134
33.273
38.393
25.0
43.0
Substance Abuse
18
45.444
4.7801
1.1267
43.067
47.822
36.0
53.0
Both
14
54.786
5.3086
1.4188
51.721
57.851
47.0
65.0
Total
50
44.600
9.0959
1.2863
42.015
47.185
25.0
65.0
Test of Homogeneity of Variances
Levene Statistic
df1
df2
Sig.
Usage Level
Based on Mean
.046
2
47
.955
Based on Median
.059
2
47
.943
Based on the Median and with adjusted df
.059
2
46.733
.943
Based on trimmed mean
.047
2
47
.954
ANOVA
Usage Level
Sum of Squares
df
Mean Square
F
Sig.
Between Groups
2848.698
2
1424.349
55.542
.000
Within Groups
1205.302
47
25.645
Total
4054.000
49
Robust Tests of Equality of Means
Usage Level
Statistica
df1
df2
Sig.
Welch
51.002
2
29.898
.000
a. Asymptotically F distributed.
Post Hoc Tests
Multiple Comparisons
Dependent Variable:
Usage Level
Tukey HSD
(I) Type of Treatment
(J) Type of Treatment
Mean Difference (I-J)
Std. Error
Sig.
95% Confidence Interval
Lower Bound
Upper Bound
Mental Health
Substance Abuse
-9.6111*
1.6880
.000
-13.696
-5.526
Both
-18.9524*
1.8046
.000
-23.320
-14.585
Substance Abuse
Mental Health
9.6111*
1.6880
.000
5.526
13.696
Both
-9.3413*
1.8046
.000
-13.709
-4.974
Both
Mental Health
18.9524*
1.8046
.000
14.585
23.320
Substance Abuse
9.3413*
1.8046
.000
4.974
13.709
*. The mean difference is significant at the 0.05 level.
Homogeneous Subsets
Usage Level
Tukey HSDa,b
Type of Treatment
N
Subset for alpha = 0.05
1
2
3
Mental Health
18
35.833
Substance Abuse
18
45.444
Both
14
54.786
Sig.
1.000
1.000
1.000
Means for groups in homogeneous subsets are displayed.
a. Uses Harmonic Mean Sample Size = 16.435.
b. The group sizes are unequal. The harmonic mean of the group sizes is
used. Type I error levels are not guaranteed.
Means Plots
Step 5: Retain or reject the null hypothesis. (This is where you report the results of your analyses t (df) = t value, p = sig. level).
Ø Based on the results above, the p-value is less than the alpha level (p < 0.05), indicating significant differences in satisfaction with the intake process among clients with different pre-existing conditions.
Step 6: Assess the Risk of Type I and Type II Error (did the data meet the assumptions of the statistic; effect size; and sample size).
Ø Assumption Check: The homogeneity of variances test (Levene’s test) is not significant (p > 0.05), suggesting that the assumption of homogeneity of variances is met.
Ø Effect Size: The effect size (Eta-squared) is not provided in the output, but it is important to consider when interpreting the practical significance of the findings.
Ø Sample Size: While the sample sizes differ across groups, the overall sample size is reasonable (N = 50).
Step 7: State your results
The results suggest that clients with different pre-existing conditions significantly differ in their satisfaction with the intake process. Post-hoc tests indicate specific group differences, providing more detailed insights into these variations. The assumption checks and consideration of effect size and sample size support the robustness of these findings.
Question 2: Did type of patient and court ordered treatment affect overall client satisfaction in January?
1. Run a Two-Way Between Groups ANOVA.
ANALYZE>GENERAL LINEAR MODEL>UNIVARIATE
2. Use NewPatient and Court as independent variables.
3. Use Overall Satisfaction in January as the dependent variable.
4. Plots are very important when looking at interactions. Whenever we see plots where the lines are not parallel, or they cross, we can be pretty sure we have an interaction. We can plot this data in two different ways (both plots will give us the same information but in different formats).
For the first plot, click on PLOT and put newpatient in HORIZONTAL AXIS and court in SEPARATE LINES, then click ADD and CONTINUE)
For the second plot, click on PLOT and put court in HORIZONTAL AXIS and newpatient in SEPARATE LINES, then click ADD and CONTINUE)
Be sure to describe what you see in the graphs.
Step 1: State the hypothesis (null and alternate)
v Null hypothesis (H0): Based on the kind of patient and court-ordered therapy, there are no variations in total client satisfaction in January.
v Alternative hypothesis (H1): Depending on the patient’s kind and court-ordered therapy, there are variations in January’s overall client satisfaction.
Step 2: State your alpha (unless requested otherwise, this is always set to alpha = .05)
Ø Alpha (α): 0.05
Step 3: Collect the data (use one of the data sets).
v Use the provided data set with NewPatient and Court as independent variables and Overall Satisfaction in January as the dependent variable.
Step 4: Calculate your statistic and p-value (this is where you run SPSS and examine your output files).
Univariate Analysis of Variance
Between-Subjects Factors
Value Label
N
Type of Patient
0
First Time
27
1
Repeat Admission
23
Court Ordered Treatment
0
No
26
1
Yes
24
Descriptive Statistics
Dependent Variable:
Overall Satisfaction in January
Type of Patient
Court Ordered Treatment
Mean
Std. Deviation
N
First Time
No
4.3571
1.27745
14
Yes
3.6154
1.26085
13
Total
4.0000
1.30089
27
Repeat Admission
No
2.5000
1.16775
12
Yes
3.7273
1.19087
11
Total
3.0870
1.31125
23
Total
No
3.5000
1.52971
26
Yes
3.6667
1.20386
24
Total
3.5800
1.37158
50
Tests of Between-Subjects Effects
Dependent Variable:
Overall Satisfaction in January
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Corrected Model
22.707a
3
7.569
5.012
.004
Intercept
625.040
1
625.040
413.856
.000
Newpatient
9.442
1
9.442
6.252
.016
Court
.731
1
.731
.484
.490
Newpatient * Court
12.018
1
12.018
7.958
.007
Error
69.473
46
1.510
Total
733.000
50
Corrected Total
92.180
49
a. R Squared = .246 (Adjusted R Squared = .197)
Profile Plots
Step 5: Retain or reject the null hypothesis. (This is where you report the results of your analyses t (df) = t value, p = sig. level).
v The p-value for the Corrected Model is 0.004, which is less than the alpha level of 0.05. Thus, we reject the null hypothesis, indicating that there are significant differences in overall client satisfaction in January based on the type of patient, court-ordered treatment, or their interaction.
Step 6: Assess the Risk of Type I and Type II Error (did the data meet the assumptions of the statistic; effect size, and sample size).
v Assumption Check: The output does not include specific information about the normality assumptions or variances homogeneity. You may want to check these assumptions separately.
v Effect Size: The R-squared value (0.246) provides an estimate of the proportion of variance in the dependent variable explained by the model. It suggests a moderate effect size.
v Sample Size: The sample sizes for each combination of factors appear reasonable.
Step 7: State your results
5. Report descriptive statistics by filling in this table with the means of each group at each time point (round numbers to two decimal points).
Table 1 Means
Type of Patient
Court Ordered (No)
Court Ordered (Yes)
First Time
4.36
3.62
Repeat Admission
2.50
3.73
Total
3.50
3.67
6. Report the assumptions tests and tests of statistical significance.
Tests of Between-Subjects Effects:
There are significant effects for the Corrected Model, Newpatient, and the interaction between Newpatient and Court. The main effect of the Court is not significant.
The R-squared value is 0.246, suggesting that the model explains about 24.6% of the variance in overall satisfaction in January.
Interaction Effect:
The interaction effect (Newpatient * Court) is significant (p = 0.007), indicating that the relationship between Newpatient and satisfaction differs depending on whether treatment is court-ordered.
Write a brief conclusion statement summarizing your results. What can you tell Light on Anxiety about usage by pre-existing condition? Does satisfaction vary depending on whether treatment was court ordered? Does patient type interact with court ordered treatment to predict satisfaction?
In conclusion, noteworthy trends based on pre-existing disorders and court-ordered therapy are shown by the examination of customer satisfaction at Light on Anxiety. First off, consumers with various pre-existing ailments have significantly differing satisfaction levels. Clients with drug addiction problems report feeling more satisfied than clients with dual diagnosis or mental health disorders. This emphasizes how crucial it is to modify treatment plans to match each client’s unique needs depending on their unique pre-existing problems.
Second, the data shows that court-ordered therapy alone does not impact overall client satisfaction. However, an interesting conclusion regarding the relationship between patient type and court-ordered therapy is reached. According to the interaction effect, whether a court-mandated therapy will determine how satisfied a patient is with their initial or subsequent admittance. Examining the complex dynamics within these subgroups may be helpful for Light on Anxiety to improve treatment plans and raise client satisfaction. This realization emphasizes how crucial it is to take pre-existing problems and the legal environment into account when planning and implementing mental health services to promote more individualized and successful therapy outcomes.