Section 1. Extremely Short Answer.
1. The manager of XYZ Group Practice sent out a patient satisfaction questionnaire to a random sample of 300 patients seen in the last 60 days. She received back 30 questionnaires. Calculate the response rate. Does she have information from a sufficient percentage of the sample to make a management decision? Why or why not?
2. State the three criteria that--by the general agreement of the scientific community or by the rules of logic--must be satisfied if a cause-and-effect relationship is to be established.
3. Which of the criteria for the establishment of a cause-effect relationship may be established by empirical evidence alone in survey research? If any are unable to be established by empirical evidence alone, how might it or they be established?
4. The objective of your study is to compare scores on the Quality of Life Inventory achieved by men who have had surgery for prostate cancer two years ago. You have two groups of men-those who are employed and those who are unemployed. There are 29 men in each group. The appropriate statistical test would be:
5. A researcher calculated the correlation, Pearson r, between two variables in order to measure the strength of the relationship between the two variables, and found it to be close +1. Describe the relationship in terms of strength and direction.
6. What statistical technique is appropriate in the analysis of survey research data when you are interested in examining the relationship between two variables measured in nominal and ordinal scales?
7. Contrast induction and deduction.
Section 2. Discussion
Compare and contrast qualitative versus quantitative studies in terms of sampling, analytical objectives, types of questions they pose, types of data collection instruments they use, forms of data they produce, and degree of flexibility built into study design.Use no more than 1 page for this question.
Section 3. Application. ( To answer Question 1 below, please fill in the chart.)
1. In a study of the frail elderly living in the community, researchers were interested in the effects of living arrangement (whether or not someone lived alone), functional status (measured by asking about ability to do six activities of daily living, resulting score from 0-6, where 6 is the highest level of impairment), gender, and self-reported health status (1=poor, 2=fair, 3=good, 4=excellent) on quality of life (measured by the Philadelphia Geriatric Center Morale Scale overall score reflecting global life satisfaction, range on this scale is from 0-22).
What is the unit of analysis? Identify the target population, the independent variable(s), the dependent variable(s). For each variable, indicate the level of measurement (i.e., nominal, ordinal, quantitative). Identify the appropriate statistical technique to analyze the effects of the independent variables simultaneously. (13 points)
unit of analysis =
target population =
statistical technique =
Variable Name Independent or Dependent Level of Measurement (Nominal, Ordinal, Quantitative)
Living Arrangement
Functional status
Gender
Self-reported Health Status
Quality of Life
--------------------------------------------------------------------------------------------------------------------------------------------------------
2. Currently XYZ Outpatient Clinic is conducting a patient satisfaction assessment by using a survey form designed by a secretary in the front office. It consists of the following questions:
Visit Date____________ Visit Time __________
Q1. Which of the following indicates your level of satisfaction with your physician?
Satisfied Very Satisfied Greatly Satisfied Extremely Satisfied Greatest Doctor in the World!
1 2 3 4 5
Q2. Please rate your satisfaction with the cost and effectiveness of the treatment you received.
Satisfied Very Satisfied Greatly Satisfied Extremely Satisfied Best Bang for the Buck!
1 2 3 4 5
Thank you for filling out our survey! Please return as soon as possible.
It is the responsibility of the billing clerk to give each patient a survey form with a request to return it as soon as possible. However, the clinic manager has noticed that while 1,000 patients come through the clinic each week, she is replacing only 800 surveys each week and is receiving only 100 responses each week, half of which are incomplete. All returned surveys are briefly reviewed by the clinic manager as they are received and filed promptly under "S" for Satisfied Patients.
As the administrative intern in this office, you have been asked to critique the current patient satisfaction assessment plan. Be sure to address survey instrument/question design (specifically discuss issues with the questions), sampling, distribution, collection, response rate, analysis and use of survey results. (20 points)Use no more than 1½ pages to answer this question.
3. Read the following abstract and answer the questions below.
Peer mentoring and financial incentives to improve glucose control in African American veterans: a randomized trial.Long JA1, Jahnle EC, Richardson DM, Loewenstein G, Volpp KG.Ann Intern Med. 2012 Mar 20;156(6):416-24. doi: 10.7326/0003-4819-156-6-201203200-00004.
Abstract
BACKGROUND: Compared with white persons, African Americans have a greater incidence of diabetes, decreased control, and higher rates of microvascular complications. A peer mentorship model could be a scalable approach to improving control in this population and reducing disparities in diabetic outcomes.
OBJECTIVE: To determine whether peer mentors or financial incentives are superior to usual care in helping African American veterans decrease their hemoglobin A(1c) (HbA(1c)) levels.
DESIGN: A 6-month randomized, controlled trial. (ClinicalTrials.gov registration number: NCT01125956)
SETTING: Philadelphia Veterans Affairs Medical Center.
PATIENTS: African American veterans aged 50 to 70 years with persistently poor diabetes control.
INTERVENTION: 118 patients were randomly assigned to 1 of 3 groups: usual care, a peer mentoring group, and a financial incentives group. Usual care patients were notified of their starting HbA(1c) level and recommended goals for HbA(1c). Those in the peer mentoring group were assigned a mentor who formerly had poor glycemic control but now had good control (HbA(1c) level ≤7.5%). The mentor was asked to talk with the patient at least once per week. Peer mentors were matched by race, sex, and age. Patients in the financial incentive group could earn $100 by decreasing their HbA(1c) level by 1% and $200 by decreasing it by 2% or to an HbA(1c) level of 6.5%.
MEASUREMENTS: Change in HbA(1c) level at 6 months.
RESULTS: Mentors and mentees talked the most in the first month (mean calls, 4; range, 0 to 30), but calls decreased to a mean of 2 calls (range, 0 to 10) by the sixth month. Levels of HbA(1c) decreased from 9.9% to 9.8% in the control group, from 9.8% to 8.7% in the peer mentor group, and from 9.5% to 9.1% in the financial incentive group. Mean change in HbA(1c) level from baseline to 6 months relative to control was -1.07% (95% CI, -1.84% to -0.31%) in the peer mentor group and -0.45% (CI, -1.23% to 0.32%) in the financial incentive group.
LIMITATION: The study included only veterans and lasted only 6 months.
CONCLUSION: Peer mentorship improved glucose control in a cohort of African American veterans with diabetes.
PRIMARY FUNDING SOURCE: National Institute on Aging Roybal Center.
Questions
1. What was the research question of this study?
2. Was this basic or applied research?
3. Who is the target population?
4. How were participants assigned?
5. What kind of research design was this?
6. What was(were) the independent variable(s)?
7. What was(were) the dependent variables?
8. Were there controlled variables? If so, what were they?
Validity Issues. Read the following scenario and answer the questions.
The Student Life Department at Upper Midwest University has decided that in order to better support new students, it will give all incoming freshman a survey about what kind of support they have in their lives to help them with the transition from being high school students living at home with their parents to being college students living on campus. The university hopes that scores from the test will help reveal students who need extra support. Mr. Jones, the director of the Student Life Department, finds four different measures of social support: the Miller Support and Affection Scale, the Latino Community Social Support, the Young Adult Social Support Scale, and the Cancer Social Support Scale.
1. Of the four measures that Mr. Jones found, which is the most likely to be one he should research further? Why?
2. Mr. Jones decides to research that scale further and finds that it has a Cronbach's alpha of 0.78. Is this score acceptable? Why or why not?
3. Mr. Jones researches more on the scale and finds out that it has been tested against other measures of social support scales for adults. What type of validity is represented by this type of testing?
4. Mr. Jones decides that he will use this scale for the survey of new freshman. Using the initial survey, the Student Life Department identifies 20 students who they feel have low social support. The Student Life Department decides to pair them up with students who have high levels of social support in a "buddy system." At the end of the year, they give the Young Adult Social Support Scale again to the 20 students to see if the buddy system worked. All of the students showed improved levels of social support. The improvement in social support levels is an example of what threat to internal validity? Why?
Displaying Data
Jamestown is a small rural town in the southeastern part of the United States. Historically, the town was mainly a farming community. However, a furniture factory has been the main source of income for most families for the last 50 years. The nearest major city is about 45 minutes away. Recently, the mayor of the town, Mr. Smith, returned from the National Conference of Mayors. Among the many topics discussed at the meeting was the growing obesity epidemic in the United States. Mr. Smith is very interested in this topic, as he has noticed that more and more people in Jamestown are overweight, including young children. Mayor Smith thought that doing something about the obesity problem in his town would benefit all its citizens. However, before he started working on programs and spending valuable community money and resources, he wanted to make sure that there was indeed a problem with obesity in his community.
When he got back from the conference, he went to the local clinic and met with healthcare providers. Also at the meeting was a clerk from the county vital records department. Mayor Smith explained his concerns to them and asked for data that would prove or disprove his suspicions. He asked the healthcare providers how many people at the clinics were overweight (according to body mass index for their height and weight). He also wanted to know their ages, whether or they had been diagnosed with an obesity-related illness like type 2 diabetes or chronic heart disease, and their race/ethnicity. From the vital records clerk, he wanted to know if obesity-related causes of death had grown over the last decade.
The healthcare providers and the clerk gave Mayor Smith the requested statistics. However, they all came on long lists of paper. You are the administrative intern, so you have been asked to prepare all the data in an easy-to-understand, straightforward fashion that he can present to the city council.
To help you prepare the data, you will need to answer the following questions related to information that was requested:
1. What classification(categorical or quantitative) and level of measurement (nominal, ordinal, interval, ratio) do age, "obesity-related illness," and, race/ethnicity each represent?
2. What type of graphical presentation would you use to show the racial/ethnic distribution of the community?
3. What type of graphical presentation would you use to show how many men and how many women were overweight by age group(grouped in ten-year increments)?
4. What type of graphical presentation would you use to show which age groups had the highest numbers of overweight people?
5. How would you show which gender had the highest numbers of overweight people?
After assembling these data for the mayor, he tells you that you are not quite done. He would like you to create some simple descriptive statistics that he can present to the city council. He would like to know the mean, median, and mode of all those individuals who are overweight. Below is a representative sample of 20 people from the data.
25 13 67 7
33 42 11 18
28 31 36 47
52 73 80 15
23 14 44 36
6. What is the mean? What is the median?What is the mode?