Sampling Techniques PDF

# Participants Needed ## How to Ensure the Sample is Representative of the Whole Target Population 1. **Identify the target population:** - Narrowly define who the research is focusing on. - Although we would prefer to generalize our results to all human beings, we need to specify who is...

# Participants Needed ## How to Ensure the Sample is Representative of the Whole Target Population 1. **Identify the target population:** - Narrowly define who the research is focusing on. - Although we would prefer to generalize our results to all human beings, we need to specify who is participating in our research. - It is very important to narrowly define the target population. For example, rather than saying that you're doing research about junk food with youth, one should specify by saying that the study is being done with 13-15 year old Maltese teenagers. - **Example:** A study about the opinion of students attending GCHSS about ask.fm is surely not including all Maltese post-secondary students' opinions and definitely not all post-secondary students' opinions in the world. 2. **When possible, random sampling should be used:** - This avoids sampling bias. - This is especially the case in quantitative studies. - A biased sample occurs when some characteristics of the population are over or under-represented thus not reflecting the real characteristics of the population. - **Example:** If I want to investigate Maltese women's view about work, if I go at Pama on a Monday at 9am, I would be excluding all other women who might be at home or at work. These women might offer a totally different view than the women I meet at Pama. ## Different Sampling Techniques We are going to discuss 7 sampling techniques: 1. Random 2. Stratified 3. Cluster 4. Convenience 5. Purposive 6. Snowball 7. Volunteer ### Random and Stratified Samples - These samples are used when we need the sample to be representative of the whole population. - They provide the most valid or credible (believable and authentic) results. - Due to the nature of these sampling techniques, they are most commonly used in quantitative studies. #### 1. Random Sampling - Since researchers usually cannot obtain data from every single person in the target population, a smaller portion is randomly selected to represent the entire group as a whole. - This is done when every person in the target group has an equal chance of being selected for the sample. - When the researched population is small, one can write the names of the participants, put them in a small bag, and pick the required number. **Example:** If the Scouts' leader requires feedback about summer camp from 10 of his 30 boy scouts, he can write the names of all the boys and pick 10 names from the bag, without looking! - When the sampling frame includes a larger number of people, a computer program (random number generator) is used to choose the sample. **Example:** If I want to select 250 women from Zejtun aged 60 - 80 years, I can randomly select them from the local council database which includes all names of Zejtun citizens. - So, a random sample is when a fragment of the studied population is indiscriminately chosen. - The problem with this method is that the people chosen might not want to participate in the study, especially when dealing with a sensitive topic. - This might create a bias in the sample, because people agree to participate may differ in various ways from those decline the request to participate in the study. - Some of the differences may be significant to the research question. **Example:** A researcher who is studying domestic violence in married couples might receive a number of declines from wives who are suffering from domestic violence because they might be afraid to disclose the problem. - **Advantages** - Ideal when dealing with large sample populations. - Avoids bias. - Everyone in the target population has equal chance of be chosen (fair). - **Disadvantages** - Requires a list of all members of the target population. - Some populations are extremely large, and we do not have all the information about each item/individual. When the case not everyone has equal chance of being selected. - Does not offer proportionally (does not offer any control on who is chosen, so you might have an underrepresented or overrepresented. *For random Sampling to be used, a list of everyone in the target group is needed.* #### 2. Stratified Sampling - Stratified sampling allows researchers to make sure that certain aspects of the target population (e.g., age, gender, religion, marital status...) which are relevant to the research are reflected in the sample. **Examples:** - In a study that studies how common it is for Maltese people to use contraceptives, AGE might be a good stratum (category) on which to divide the sample. - In another study about views on sharing work-life responsibilities, GENDER might be an important stratum. - In a stratified sample, after establishing the strata needed to be used, a random sample is used that reflects the proportion of those characteristics in the target population. - The ratios used for each strata in the research reflect ratios in real life. - Stratified Sampling is a very useful tool when carrying out a quantitative study. - A similar sampling technique used in qualitative studies is called quota sampling. - In quota sampling, after establishing strata, participants are chosen in a non-random way. **Example:** Name of study: The different views of 1st year and 2nd year GCHSS students about the school administration personnel. | Year | Number of Students | Required Participants | |--------|--------------------|-----------------------| | 1st | 1000 | 135 | | 2nd | 500 | 65 | - When using stratified sampling, students are then randomly selected until the numbers of that strata are reached. - While in a quota sampling, the researcher might stay near the school gate and choose participants who conveniently show up, until the right number of participants is acquired. - **Advantages (stratified sampling)** - Guarantees proportionality. - Hence, it is more representative of the whole population as we have control on the number of people that are chosen from each stratum. - Stratified sampling prevents bias as random sampling is also used in the process. - More accurate than random sampling. - Correlations and comparisons are more easily made between strata. - **Disadvantages (stratified sampling)** - More time consuming than random sampling. - The proportions of the strata must be known and accurate if it is to work properly. - Accurate up to date population data may not be available and may be hard to identify peoples age or social background effectively. ### Other Sampling Techniques: - Used in qualitative studies or in quantitative studies when lists are not available. In this case, samples are gathered in a process that does NOT give all the individuals in the population an equal chance of being selected. - As these samples are **not truly representative**, they are less desirable than random and stratified samples. - However, a researcher may not be able to obtain a random or stratified sample, or it may be too expensive. - A researcher may not care about generalizing to a larger population, especially when opting for qualitative methodology. - Since they do not truly represent a population, we cannot make valid **assumptions about the larger population** (we cannot generalize results). #### 1. Cluster Sampling - Carried out by dividing the studied area in clusters, selecting at random the desired number of clusters and then randomly selecting participants from those chosen clusters. - This is used when it is impossible or not practical to include everyone in the target population. - Used especially when population members are widely scattered geographically. **Example:** It would be impossible to interview all pastors in the United Kingdom. In this case, the target population is divided into a number of small subdivisions (clusters), for example, by region, and then some of these clusters are randomly selected. - So if the researcher needs 100 pastors in all, this amount will be chosen randomly from the chosen clusters only. #### 2. Snowball Sampling - Carried out by asking participants to invite other people they know who fit the criteria to participate in the research study. - Usually carried out when lists of target population are unavailable or participants might not wish to be identified (e.g., women who did an abortion, children who were raped, prostitutes…). **Examples:** - Samuel is doing a study with persons who used illegal drugs. After finding a person from the target population, he asks the participant to invite other people he knows who went through the same experience. - Anna is doing research about Maltese women who were fooled by an online scam. She already knows two individuals but since she needs 8 in all, she asks them whether they have met other women who had to endure this experience. #### 3. Purposive Sampling - Usually when purposive sampling is used, sample size is small. - The goal is not to choose participants randomly to make generalisations, but to focus on particular characteristics of a population. - Makes much more sense to use it in a qualitative study where the researcher needs to choose particular people on purpose. - Individuals who are expected to offer the most detailed and appropriate information for the study will be approached and invited to participate. - This leaves room for a lot of subjectivity on the researcher's side which might cause a sampling bias. - Hence, the researcher should acknowledge and justify why he used this sampling technique. **Examples:** - **Quantitative study:** A University student is distributing a questionnaire about branded handbags and accessories, so she targets people in the streets and picks women who are 'dressed nicely'. - **Qualitative study:** Another researcher might be interested to learn about the lived experience of people suffering from diabetes. It might be very difficult to go around the general population and ask people if they are suffering from this condition. Instead, the researcher could attend a support group for diabetic individuals. - Since these people were comfortable to talk to other sufferers, they might be more likely to talk to the researcher and provide valuable information. The researcher in this case may invite those individuals who are more willing to participate and share experiences during meetings held. #### 4. Volunteer Sampling - This is used when researchers request for participants via adverts, social media, emails, on a newspaper noticeboard and other means and participants volunteer on their own. - The researcher might then only approach suitable individuals for the study - **Advantages:** - If advert reaches a large number of people, it can be a quick and convenient way to recruit participants. - If a large number of participants is included, results may be more in-depth and accurate. - **Disadvantages:** - People who did not see the advert were excluded. - Volunteers might be significantly different from those who did see the advert or saw it and refused to participate. - Some people might join or volunteer are promised a reward. - Likely that the sample is not representative. #### 5. Convenience Sampling - In this case participants are not randomly selected from target population, but the researcher uses any individuals that are available and willing to participate in the study. **Examples:** - Handing out questionnaires to every person walking in Baystreet at Paceville. - Surveying relatives, friends, students, or colleagues that the researcher has regular access to. - **Advantages:** - It is cheaper and easier, since, rather than sampling an entire country when using simple random sampling, the researcher can allocate his limited resources to the few randomly selected clusters. - The researcher can also increase his sample size since the population chosen is more accessible. - **Disadvantages:** - Part of the population is deliberately excluded from taking part since not all clusters are chosen. Excluded portions might have different opinions than chosen portions. - This technique is the least representative of the population as the tendency of individuals within a cluster is to have similar characteristics. - Researcher might have an overrepresented or underrepresented cluster which can influence the results of the study. **Important Considerations:** - **Target Group:** A critical term, signifying the specific population being studied. - **Sample:** A subset of the target group chosen to represent the entire population. **What is Sampling?** - A sample is a portion of the population that is being studied. - The chosen sample should be representative, meaning it should represent the characteristics of the target group so that the obtained results from the sample can be generalised to the whole target group. - This is especially important in quantitative studies, where the researcher needs to choose an appropriate sampling strategy to obtain a representative and statistically sound sample of the whole target population. **Advantages of Sampling:** - **Quick & Practical:** Efficient for gathering data. - **Economical:** Cost-effective compared to studying the entire population. - **Easy:** Participants are readily available, streamlining the process. **Disadvantages of Sampling:** - **Excludes Large Proportion:** Results might not reflect the entire population. - **Potential Biases:** The sample might not accurately reflect the target group, leading to inaccurate conclusions. - **Difficult to Generalize:** Results might not be applicable to the entire population due to the limited sample size. **Key Points:** - **Best or Exam Only:** In quantitative studies, choose only one sampling technique. - **Combined Techniques:** In certain studies, the researcher might need to combine two or more sampling techniques together. - **Qualitative Studies:** The goal is to obtain rich and detailed information, not to generalize the results. - **Volunteer Samples:** Useful especially when participants are not easy to find. **In Summary:** - Sampling methods are crucial for conducting research effectively. - Understanding different sampling techniques is essential for choosing the most appropriate method for your study. - Choose a method that will provide a representative sample, reflecting the characteristics of your target population. - Ultimately, your chosen sampling technique should aid in obtaining accurate, reliable, and valid results.