Conducting Educational Research
Step 7: Write Population, Sample, and Sampling Technique Sections

Since the target population and sampling technique have already been identified, this step simply consists of writing the relevant sections for the research report. Recall that the researcher has an important responsibility to systematically report on the methods of a research study so other researchers can replicate the work. Therefore, it is best to give more details than is thought necessary when writing this section. In addition to allowing other researchers to replicate the study, accurately reporting all details regarding the participants in the study (including the population, sample, and sampling technique) is also necessary for the following reasons:

Write Population Section

The population section should describe relevant demographic (biodata) characteristics of the population that were identified in Chapter 11. This might include gender, age, type of school (public/private; urban/rural; day/boarding), special educational needs (e.g., pupils with a disability), etc.

Write Sample Section

The sample section is one case where writing a dissertation typically differs from APA guidelines for writing journal articles. According to APA guidelines, the major demographic characteristics of the sample must be described, including sex, age, and race/ethnicity as well as socioeconomic status and disability status where possible and appropriate. This means that the researcher must tell how many participants were male/female, the average age of the sample, and describe the ethnicity of participants. Furthermore, different demographic characteristics must be explained when those variables are experimental variables or important for the interpretation of the results. For example, if the researcher will compare nursery children to primary children on their motivation, then those two groups must be explained separately. How many males/females were in nursery school? How many were in primary school? Other characteristics such as national origin, level of education, health status, and language preference can help give readers a better understanding of the sample. Finally, in experimental, quasi-experimental, and causal-comparative designs, state the total number of participants in the study and the number assigned to each experimental condition or group.

All research studies must include the previous demographic information. However, completing a dissertation is different than writing an empirical paper for publication. The sample section of a dissertation is typically written before the sample has been properly identified. Therefore, it is oftentimes not possible to thoroughly describe the sample in first drafts of a dissertation. The researcher should work with their supervisor to determine how much information needs to be included in the sample section before the study has been completed. After the study is finished, the researcher needs to determine whether the relevant demographic information should be included in the sample section or in the beginning of the Results section.

There are two other points that should be considered when writing the sample section. The first is missing data and the second is ethical considerations.

Missing data. All studies will have missing data and/or participants who drop out of the study. This is perfectly normal, so researchers have developed acceptable methods for handling missing data. In absolutely no instances should a researcher make up data. Instead, the researcher should educate themselves about acceptable practices for overcoming the problem of missing data.

There are two types of missing data. First, a participant might have just skipped one item on a questionnaire. Strategies for overcoming missing data on just one or two items will be described when discussing data analysis in Coding Data from the Instrument. The second type of missing data is when a participant is missing a score for one or more variables in the study, perhaps because they were absent on the day of testing or they skipped one page of the questionnaire.

There can be two possible outcomes for missing data on an entire variable. In some cases, enough information is still available for the participant to be included in the sample. For example, if a researcher is comparing boys' and girls' achievement in English, Maths, and Science, a student may have a score in English and Maths but not Science. In this case, the researcher can use that student's English and Math scores, but they will be considered missing in Science. It is best to report this type of missing data in the results section. In the table with the mean scores for Maths, English, and Science, you must also report the sample size (symbolized by "N") for each of the mean scores. Perhaps there were 53 Maths scores, 52 English scores, and 47 English scores. These sample sizes should be accurately reported in a column in the table of mean scores in the results section.

In other cases, missing data may exclude the participant from the study entirely. For example, in a pre-post test study, 100 participants completed the pre-test, but only 86 participants completed the post-test. In this case, the 14 individuals who did not complete the post-test have to be removed entirely from the analysis. This information needs to be reported in the sample section, as well as a brief explanation of why they did not complete the post-test. "One hundred participants completed the pre-test, but only 86 completed the post-test. Of the 14 participants who did not complete the post-test, 9 were absent from school on the day of the post-test whereas 5 others transferred to different schools during the time of the experiment."

There may be other participants who have available data but are excluded from the analysis for various reasons: perhaps their responses on the questionnaire were not clear, or they fell outside the target population (maybe only people ages 18 and older were identified as the target population, but a few 17 year olds completed the questionnaire), or the participants did not attend enough of the treatment sessions. All of these details must also be reported in the sample section: How many participants were excluded from the analysis and for what reasons?

Finally, researchers must accurately report the number of participants in a sample. While researchers should work hard to get as many questionnaires returned as possible, it is practically impossible that all questionnaires will be returned. Perhaps the researcher distributed 100 questionnaires, but only 92 were returned. The sample size for this study is 92, not 100. Researchers know that not all questionnaires will be returned.

Ethical principles and code of conduct. As stated in Ethical Considerations, researchers must follow standard ethical principles when interacting with human participants, including voluntary participation in the research, informed consent, and confidentiality and anonymity where appropriate. A statement in the Sample section should state that APA's ethical standards were followed.

To summarize, the Sample section should include:

  • Number of participants broken down by major demographic characteristics (e.g., age, grade, gender, race, language, socioeconomic status) and the number of participants assigned to groups or treatments
  • Describe any missing data or excluded participants and why
  • Statement of ethical treatment

Here is a sample Participants section from an empirical paper (Korb, 2010).

All participants selected for this study resided in the middle-belt of Nigeria. Two samples were obtained. The first sample included 42 practicing teachers. These teachers attended a four-week in-service training workshop on general teaching skills that culminated in earning a certificate from a university. (The workshop was unrelated to literacy development.) In this sample, 83% taught at government schools and 7% taught at private schools (10% did not respond). In terms of qualification, 41% earned a National Certificate of Education (the basic qualification for teaching at primary school), 27% completed a university bachelor's degree, and 31% earned a postgraduate degree. The mean number of years taught was 10.6 years (SD=6.2 years).

The second sample consisted of 171 pre-service teachers who were in their first year of teacher training at the university. Three different questionnaires, one of which was the questionnaire for this study, were randomly distributed as course credit to the 550 students enrolled a general education course. Because students randomly received different questionnaires, the participants in this study represent a random selection of all students enrolled in this course. Of the pre-service teachers, 13% taught before attending university, with an average of 2.7 years of teaching experience (SD=3.80 years). The gender and age of the two samples are described in Table 1.

No participants were excluded for any reason. All participants were voluntary and received informed consent, in line with APA's ethical standards.

Write Sampling Technique Section

Recall that the research design section should be written in terms of both the theoretical and the practical: the theoretical includes a general description of the research design as described by research methodologists, whereas the practical describes how the core components of the research design are implemented in the study. (Click on Research Design for a refresher if needed.) Likewise, the sampling technique should also be written in terms of the theoretical and practical. First, the sampling technique should be explained. If simple random sampling was chosen, what does simple random sampling mean? Once the sampling technique has been theoretically explained, then explain how the sampling technique was practically applied to the particular study: how was the list of the target population obtained? How were members on that list randomly sampled? An example Sampling Technique section is written below.

Multistage sampling was chosen for the study. This type of sampling technique consists of sampling at multiple stages (Singleton & Straits, 2010). In this study, the unit of analysis is students. First, the students were clustered into states. One state from each geopolitical zone was randomly selected using the hat and draw method. In the second stage, an LGA within each state was randomly selected, also using the hat and draw method. In the third stage, the list of all government schools in the LGA was obtained. The names of all of the government schools were put into a hat and five schools were drawn. In the final stage, the list of students in JS2 was obtained in those five schools. All JS2 students in the five selected schools were purposively chosen to participate in the study.

Final consideration. If the word random appears anywhere in the sampling technique section, that means that participants had to have been selected purely by chance using standard random procedures developed by researchers - not because of any other reason than that they were selected out of a hat or by a random number table. If there was anything else that influenced the selection of specific participants, then the word random must be canceled from this section.

Finally, reporting empirical research in a journal article, technical report, or book chapter is different than writing a thesis. If you are not writing a thesis, then the population, sample, and sampling technique sections are typically combined into one Participants section. In the participants section, briefly identify the target population, then thoroughly describe the sample and how the sample was chosen, using the same guidelines presented above.


Return to Educational Research Steps

Copyright 2012, Katrina A. Korb, All Rights Reserved