solved 1 short paragraph describing the limitations of these findings. Â

1 short paragraph describing the limitations of these findings.  
As you’ve learned in class, descriptive statistics are used to describe and summarize the visible characteristics of a sample and we have practiced using inference and estimation to determine if there are statistically significant differences between our sample statistics and the population parameter. Finally, you have learned how to conduct a t-test and analysis of variance to learn to compare means.   
For this group assignment, you will use univariate and bivariate descriptive statistics, conduct a T-test, interpret p-values and confidence intervals and conduct an ANOVA to describe and summarize features of the Health and Retirement Study. 
Step 1: Select 1 health equity-related question that can be answered with our 2008 (variable names RAND: r8*;  Psychosocial and Lifestyle questionnaire: KLB*) HRS data from the following list: 

Are there gender (ragender) and race/ethnic (race4) differences in Life Satisfaction (KLBQ03a-e) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Hopelessness (KLBQ19L-o) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Loneliness (KLBQ19a-c) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Religiosity/Spirituality (KLBQ28a-d) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Perceived Everyday Discrimination (KLBQ30a-e) among older adults? *(items in scale must first be reverse coded)
Are there gender (ragender) and race/ethnic (race4) differences in Purpose in Life (KLBQ35a-g) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Major Experience of Lifetime Discrimination (KLBQ36a-g) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Life Time Traumas (KLBQ37a-g) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Life Time Traumas before 18 (KLBQ37a-d) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Stressful Life Events (KLBQ38a-f) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Ongoing Chronic Stress (KLBQ40a_a-h) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Anxiety (KLBQ41a-e) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Depression (r8cesd) among older adults?
Are there gender (ragender) and race/ethnic (race4) differences in Chronic Work Discrimination (KLBQ49a-f) among older adults?

Step 2: Identify and recode the variables in the dataset that you will use to answer your chosen question. You’ll begin by determining which are your independent variables and which is your dependent variable, and then finding them in the codebook and the dataset. You will need to follow the instructions in the codebook for the Psychosocial Leave behind Survey to create the scale appropriate for your research question. Note: Some scales will be the sum of the variables in the scale, sum will average across variables to create the scale. Please make sure you note whether you summarized or averaged across the variables in your scale; present an alpha for scales that are averaged).
Independent Variables: Race (a1RACE) & Gender (a1GENDER)
Dependent Variable: Lifetime Traumas (a1_rS_LTT) <- To scale this, we took the sum of the variables.  Step 3: Use univariate descriptive statistics to summarize the characteristics of each of your variables, including the mean, standard deviation, and range for your continuous variables and the frequency and percentages for your categorical variables.    Descriptive Statistics N Range Minimum Maximum Mean Std. Deviation a1_rS_LTT 6941 7.00 .00 7.00 1.2320 1.19247 Valid N (listwise) 6941 Step 4: Use bivariate descriptive statistics to summarize mean gender and race/ethnic differences in your outcome variable.  To do this, you will produce a means table by 1) gender and 2) race/ethnicity using the Compare Means command (include standard deviation, standard error, range). Hint: you should have measures of central tendency and dispersion for men and women separately and for each race/ethnic group separately.  a1_rS_LTT  * ragender: r gender a1_rS_LTT   ragender: r gender Mean N Std. Deviation Std. Error of Mean Range 1.male 1.2908 2930 1.22122 .02256 7.00 2.female 1.1890 4011 1.16930 .01846 7.00 Total 1.2320 6941 1.19247 .01431 7.00 a1_rS_LTT  * a1RACE a1_rS_LTT   a1RACE Mean N Std. Deviation Std. Error of Mean Range white 1.2260 5535 1.17176 .01575 7.00 black 1.2503 891 1.23761 .04146 7.00 hispanic 1.2641 515 1.32771 .05851 6.00 Total 1.2320 6941 1.19247 .01431 7.00 Step 5: Calculate and interpret a 95% confidence interval around the means for each race/ethnic and gender group and produce a Simple Error Bar graph that shows the means and the 95% confidence intervals around the means for each group (one graph for gender differences; one graph for race differences) Analyze < Descriptive Statistics < Explore < Dependent variable: BMI < Factor List: RACE  < Under Display: select “Statistics” < Statistics  command < Check “Descriptives: Confidence Interval for Mean: 95%” < Continue < OK Graphs < Chart Builder < Choose from: Bar < Simple Error Bar Mean < put race as your x axis (along the bottom) < put BMI as the y axis (along the side) < Errors Bars Represent: Confidence Intervals, Level: 95 < OK Step 6: Conduct an independent sample t?test (two-sample t?test for significance of difference between two means) to determine whether the difference in the mean of your dependent variable between men and women is statistically significant (i.e., different from 0). Reach a conclusion.  **because p-value is <0.001, equal variances are not assumed and we can conclude that there is a significant difference b/t genders and the lifetime traumas that they face** **Note: Be sure to use the “Levene’s Test” p-value/sig to determine if you should “equal variances assumed” or “equal variances not assumed” output.  Step 7: Conduct an ANOVA (F-test for significance of difference between more than two means) to determine whether there are race/ethnic differences in your dependent variable. Interpret the F-value in terms of between/within group differences. What can we conclude about race differences in your dependent variable? **From the ANOVA, based on the p-value of 0.697 (>0.005), we can conclude that there is no significant difference between races and the experience of lifetime traumas**
Step 8: Develop a final product that includes the following: 

Group member names, date, title of the assignment
Your chosen research question from the options above
Original variables included in analysis. Be sure to include the full information on the variables from the codebook, including the survey question that was asked of the participants. You may copy/paste this but it should be formatted like the rest of the document.
Brief description of any re-coding of variables and why you re-coded them that way. Brief description of how you created the scale or count variable that is your outcome/dependent variable.
Brief description of your approach to analysis, including your steps for producing the T-test and F-test. 
Tables, charts, and/or graphs presenting your findings in a way that makes sense and authentically presents what you found (remember what you learned about misleading graphs!). Include a Table that summarizes all the following information in one place:

Gender and race/ethnic differences in (dependent variable) among older adults

n
Mean
SE
CI
T or F value
P-value
(T or F test)
Gender
    Men
2930
1.2908
.02256
1.25 -1.34
8.996
0.003
    Women
4011
1.1890
.01846
1.15 -1.23
8.996
0.003
Race/Ethnicity
    White
5535
1.23
.01575
1.20 – 1.26
0.361
0.697
    Black
892
1.25
.04146
1.17 – 1.33
0.361
0.697
    Hispanic
515
1.26
.05851
1.15 – 1.38
0.361
0.697

1-3 short paragraphs describing what you found, i.e., describing the outcomes/findings of your crosstabs, chi-square 1.33test of independence, T-test and ANOVA.

Based on the results of the T-test, p-value is <0.001, equal variances are not assumed. We can conclude that there is a significant difference between males and females and the reported lifetime trauma that they faced. Based on the  ANOVA testing, there is a P-value of 0.697 which is greater than the significance level 0.05. Therefore, we can conclude that there is no significant difference in mean between race/ethnicity and Lifetime Traumas. Because the results of the ANOVA indicated that there was no significant difference, there was no reason to check the Scheffe test of multiple comparisons; however, reading over did confirm that there was no significant difference between races and lifetime traumas.  1-2 short paragraphs describing the implications of your findings (the “So what?” for public health). Find ONE study in APA format and describe how they support or disprove your findings. In what ways is your study design and sample similar or different from these studies?  According to our findings, trauma does not discriminate by race or ethnicity, but does occur statistically more often to women when compared to men. This is probable because women are more likely to experience lifetime traumas, such as assault, because of perceived vulnerability, in addition to the fact that women are statistically more likely to report these occurrences.  The results should be used to justify a refined study of lifetime traumatic events for women stratified by race. From those results a public health need for women of race or women in general can be assessed and addressed.  In a similar, yet more in depth, lifetime trauma and PTSD research by Roberts et al., from the NESARC (National Epidemiologic Survey on Alcohol and Related Conditions) second wave follow up interviews in the year of 2005, they found different findings than our own. Apart from the prevalence of PTSD development in those exposed to trauma, their findings revealed higher rates of trauma exposure in whites than any other group. 86% of the White subjects had traumatic exposure and 76% for Black, 66% for Asians, and 68% for Hispanic subjects.  The information is found in Table 1 of the research. However, there were strong statistical differences between traumatic exposure origins. For example, the Black ethnicity/race subjects had higher rates of domestic violence exposure and Asians had higher rates of war trauma exposure.  Some differences of the NESARC review from our own study was the sample size of over 34,000 people, compared to our 7030 subjects, accounting for the Asian/ Pacific Isalnder/ Hawaiian/ non-hispanic population, 22 questionnaire screening for traumatic events, and the age group was 18 years and older. The study was further designed to look at the likelihood of developing PTSD (Post-Traumatic Stress Disorder) by race and gender. Furthermore, the research studied the likelihood that different races and genders sought medical treatment for PTSD.  The race ethnic differences study by Roberts, A.L. et al. shows a different disparity of traumatic event exposure of White and Black people being statistically similar to each other but significantly higher than hispanic and asian communities in America. The research doesn’t support our findings of having no relationship of traumatic exposure, but demonstrates a different aspect on where the statistical disparity of traumatic exposure may be present between race and gender. Roberts, A. L., Gilman, S. E., Breslau, J., Breslau, N., & Koenen, K. C. (2011). Race/ethnic differences in exposure to traumatic events, development of post-traumatic stress disorder, and treatment-seeking for post-traumatic stress disorder in the United States. Psychological medicine, 41(1), 71–83. https://doi.org/10.1017/S0033291710000401 1 short paragraph describing the limitations of these findings Questionnaire limited. Some traumatic events may have been overlooked because they were not asked.  Some individuals may have declined to answer because they didn’t want to recall or explain/provide info for “when?” - possible fear of reprisal from responses.  No data for other racial/ethnic groups (Native Americans, Asians, etc) Group member contribution statement: 1 paragraph explaining what each group member contributed to the project and product submitted. At least one sentence per group member.  Tory picked the question for the group assignment and he identified the independent variables and dependent variables. Torey did research to find studies similar to the one we worked with here, and provided information on their findings. Madeleine ran SPSS and collected the data for every step, assisted in the interpretation of the findings, and wrote the introduction discussing our variables and the recoding of them. John and Sue helped with the descriptions of the findings of the T-test and ANOVA. We filled out the race/ethnicity difference (variables) tables together. Klara and Lily helped with the limitation of these findings. Step 9: Submit your work as one document in Word or PDF form (no “Pages”) as a group via Canvas.  There should only be one submission per group.  Step 10: Individually provide a review of each group member’s contribution using the provided form on Canvas. Each group member will complete and submit their assessment individually.  Considerations for interpreting your findings:  Differences in health or other outcomes by specific demographics (e.g., race/ethnicity, parental education, age, gender) should not be interpreted to mean that there is anything inherently different about people by demographic factors. Rather, people in different social positions or with different social identities (e.g., male vs. female; Black vs. white vs. Latinx) may have different conditions in their social and physical environments as well as different access to opportunities to be healthy. Be sure to keep this in mind when you interpret your results. Thus, you should focus on the implications of the results for health promotion or health equity, rather than trying to identify a “cause” or specific reasons for what you found. You would need to leverage theory (at a minimum) and ideally longitudinal data to be able to make any concrete statements about why you see what you see in the data. Since we are not asking you to bring in theory or look at changes over time, please do not discuss “causes” of what you find!