Question

Please read these instructions and answer the questions below.  

Assume that you are a social work practitioner at an agency that provides mental health support services to military veterans. The agency currently serves 396 veterans. Your supervisor asks you to develop a questionnaire and administer a survey to these clients to determine their satisfaction with services and how the agency might improve services. There is a particular concern related to complaints about services to transgender and women clients.  

Your supervisor said:  

“Be sure to include a good representation of our clients by gender! Don’t worry about combat experiences! And we have enough money for participation gift cards for up to 100 clients”.  

	Men	Women	Transgender/Non-binary
Vietnam War	60	20	4
War in Iraq or Afghanistan	160	80	6
Non-combat	40	20	4
Total	260	120	14

Table showing the numbers of clients served by gender and history of combat experiences.

1. Describe how you could go about drawing a convenience sample of veterans/clients to survey.  

2. Describe how you could draw a systematic random sample of these veterans to survey?  

3. Describe how you would draw a disproportional stratified random sample (representing gender) of the veterans to survey? Assuming you have the capacity for 100 participants or so, what would you do?  

4. Explain the motivation for drawing a disproportional stratified random sample—why might you want to do it?  

Answer 1

1) As the mental health varies according to gender (and also according to the discipline served in but as given in question, we ignore it); the given data is heterogeneous. here convinience sampling coukd be used for pilot survey according to which of the veterans is willing to participate in the survey and the first 100 willing participants will give us cinvinience sample.

2) we can number each of the veterans from 1 to 394 (total number of veterans including men women and transgenders are 394). we want to select 100 veterans from a population of 394 veterans. 394/100=3.94.

the population is not evenly divisible.so, If you take every 4th veteren, 100*4=400, so there is a risk that the last 2 veterens chosen does not exist. On the other hand, if you take every 3rd veteren, 100*3=300, so the last 94 veterens will never be selected. The random starting point should instead be selected as a non integer between 0 and 3.94 (inclusive on one endpoint only) to ensure that every veteren has equal chance of being selected; the interval should now be non integral (3.94); and each non integer selected should be rounded up to the next integer.

3) As the number of veterens sampled from each gender which is a stratum is not proportional to their representation in the total population, Population elements are not given an equal chance to be included in the sample of size 100. hence we go for disproportional stratified sampling. We give(260/394)% chance to males to be in the sample; (120/394)% chance for females to be in sample and (14/294)% chance to transgenders to be in the sample.