The approach of studying entire population or just a sample of that population isn’t merely choice of a researcher but is dependent on many other components. Both of these approaches have their own benefits and drawbacks. Studying entire population is time consuming, expensive, convoluted and may lack all the resources but it is also more extensive. By studying entire population; we are more likely to get a deep insight into the research we are interested in. With right amount of funding, time and human resources it is possible to take this approach. An entire population research generally consists of large sizes of populations with some common or similar characteristics. While entire population approach is rarely taken, it can be more useful to elucidate certain research. It is difficult than studying population samples, but entire population technique can be used in the research where population itself is relatively small. In that scenario, the variable we are analyzing can be studied meticulously. Other than one similar characteristics of each variable; we can succeed in finding other components which are accountable for our main research topic. This can help us have a better understanding of whole population of our interest.

For instance, the study concentrated in Wake County, NC (2011), the research question being, “After death of an infant child, what are the psychological challenges faced by the parents?” According to NC Department of Human and Health Services, State Center for Health Statistics 2011 reports; in Wake County, total 91 infants died in that year. This is comparatively a small number where it would be advisable to take a total population research approach.  Also, this is a very important study as all the variables/parents share the similar characteristics of losing their beloved infant children. Although, the circumstances under which children died are different for each variable/parents, by taking population research approach, we can find out more about how this unfortunate episode affected their lives. All the parents of deceased 91 infants come from different socio-economic background, race, family dynamics and social class. By thoroughly investigating these characteristics we can find out what challenges each of the parents had to endure to cope up with the tragedy of losing a child.

It is evident in such researches where we need to find all the elements, it’s better to take an entire population approach.  As every parent of the 91 infant children went through same traumatic experiences but the causes and effects for each case were different. Since, 91 is small population, it will be less time consuming and it will be easier to find most of the parents. If the same research is done with sampling approach, it will be improbable to know all the aspects of the aftermath of losing a child. Sampling can be biased and less efficient in providing the needed awareness of the situation. The resulted data will be important to understand the grieving parents and psychological effect of the incident better but sampling error can be obstacle in making any prediction.

It is rare that researcher have an opportunity to do entire population research because of scarcity of resources. In most cases, sampling technique is used to produce quicker results. It is the most common approach taken by researchers to study the population of their choice. Sampling saves time and cost, it is reliable when studying large population and it takes less efforts to reach to the judgment.





For example, “How satisfied are iPhone 6 customers with their purchase?” Millions of people bought iPhone 6. It’s impossible to reach all the people and find out how satisfied they are with their purchase. Here, sampling 100 people and asking them the same question will yield better results. It will be faster and cheaper too. Questions like these involving large populations lend themselves to sampling technique.

I have not had an opportunity to experience the whole quantitative research process myself. If any of you have done any quantitative research in the past using any of these approaches, what were the difficulties faced by you?


Just for fun,





 North Carolina Department of Human and Health Services, State Center for Health Statistics 2011 report.

CHAPTER 1- Social Statistics for a Diverse Society, Chava Frankfort-Nachmias, Anna Leon- Guerrero





7 thoughts on “SOCY508_Population

  1. Thanks Rucha! Very informative. In my professional life I am responsible for sending out annual surveys to alumni. My target population is all nursing students who graduated in selected year. My biggest problem I face is response rates, largely due to lack of contact information and therefore alumni do not receive the survey. Since an important piece of population study is to ensure everyone in the target population receives the measure (i.e., survey), I suppose this would make my research to find levels of program satisfaction of alumni is actually a sample study??

  2. Dear Sam,

    From the information you provided, I would say so. If only few alumni are responding to the survey, you will be forced to make a judgement of how satisfactory program or coursework is based on those few responses. I am not sure if this happened to you, but I would also like to add that some alumni who receive the survey may just forget or not find it important to respond too. I say that because I have seen people doing it. In my opinion, there should be a provision to make students give their opinions about program or coursework right after their final exam while they are still present in the school. I think that will definitely receive more feedback. That too hassle free :). What are your thoughts about that?

    1. I would agree! Forgetting about and/or not having the time are barriers to receiving feedback from alumni. Good call-I also administer an end of programs survey seniors take in their final semester. We get better response from that-nearly double.

  3. I think you bring up an important point in saying that a population can be relatively small. We usually think of populations as large, but that’s not necessarily the case. The researcher conceptualizes and defines the population of interest. Another researcher may or may not agree with these definitions. You discuss a study of parents coping with infant deaths in one county during one year. These 91 deaths may be defined by researcher #1 as the population to which he/she would like to generalize, so the researcher would attempt to study the circumstances of the entire population of all 91 deaths. Assuming data can be collected on all 91, the entire population would be covered. However, researcher #2 might observe that this county is only one of many in North Carolina and, relative to many others, is highly urbanized, and possibly unrepresentative of general statewide conditions. Therefore, researcher #2 might think that all counties in the state would be the relevant population for that particular research question. (And so on up the geographic ladder!) A particular researcher’s focus and questions may define a population as large, small, or in-between. In applied research, the population may be defined by the unit that exercises control. For example, if Wake County officials can offer special programs and resources to bereaved parents, then Wake County infant deaths during one year may define the appropriate population for study.

  4. Dr. Honnold,
    Thank you very much for giving me a different perspective. I agree, it depends on the researcher that how to pursue this research and on what level. I also feel that Wake County officials would concentrate on their County’s population for this particular research topic to offer special programs or counselling to grieving parents to cope up. Parents of 91 deceased children will be easily reachable and may offer some solutions on how to deal with these type of cases in the future. Also, if grieving parents can offer more insight into what all they socially need, to develop a coping mechanism.

  5. I think you chose two excellent examples of ways in which one of the options will work, but the other wouldn’t be as desirable. I can’t imagine doing a population survey on satisfaction with the iPhone 6. It would be a nightmare.

Leave a Reply

Your email address will not be published. Required fields are marked *