Janine McElroy Ben Tieniber Chris Herr Selecting a Sample Janine McElroy Ben Tieniber Chris Herr

Overview Today we will review the different methods of selecting a sample Sampling in quantitative research and how to use random and nonrandom sampling techniques Sampling from qualitative research and how to use purposeful sampling

Sampling in Quantitative Research Why use a sample rather than an entire group? The group is too large Geographic reasons If sample is well selected the results will represent the entire population

Example I want to find out if high school seniors at WHB feel college ready as they near graduation Not going to spend 30 minutes with 250 students Take 10% of a correct sample will give same results

How to select a representative sample using random sample techniques ? Four types of random sampling Simple random sampling Stratified sampling Cluster sampling Systematic sampling Each of these practices requires the researcher to Identify the population Determine required sample size Select the sample

Simple Random All individuals have equal chance of being selected Completely random High probability of getting a representative sample Example: Place all 250 names of seniors in a hat and pull 25 Use a table of random numbers

Stratified Sampling Guarantee to represent desired subgroup Strategically selecting participants from each subgroup Random sample from each subgroup Example: Rank the seniors from valedictorian to last student and divide the class into 5ths. Randomly select from each group

Cluster Sampling Groups, not individuals selected When unable to attain a list of names of a population Clusters identified as schools, classrooms, hospitals Example: Look at the master schedule for senior only classes, break up into AP, Regents and Special Education classes and chose classes until I reach desired sample. My 25 student sample not good representation on cluster sampling

Systematic Sampling Not used often Sampling with every Kth individual List includes all names of individuals in population K is the variable determined by dividing the number of individuals on list by the desired sample Not an independent chance, after the first name selected the rest of the names are automatically determined Example: If K=5, taking every 5th name on the list Interviewing seniors with a class of 250 students and a desired sample of 25, K=10

Determining Sample Size The sampling question most asked by beginning researchers is, “How large should my sample be?” And the answer is LARGE ENOUGH. Some general rules: The larger the population size, the smaller the percentage of the population required to get a sample. For smaller populations, say, N=100 or fewer, there is little point in sampling; survey the entire population. If the population size is around 500, 50% should be sampled. If the population size is around 1,500, 20% should be sampled.

Selecting a Nonrandom Sample Nonrandom sampling is the process of selecting a sample using a technique that does not permit the researcher to specify the probability, or chance, that each member of a population has of being selected for the sample. Nonrandom sampling approaches include: convenience sampling, purposive sampling, and quota sampling.

Convenience Sampling Also known as accidental sampling or haphazard sampling The process of including whoever happens to be available at the time. Two examples are seeking volunteers and studying existing groups “just because they are there.” For example: You’re stopped at the mall by someone who wants your opinion of an event or of a new kind of cookie

Purposive Sampling Also referred to as judgment sampling and is the process of selecting a sample that is believed to be representative of a given population Sample selection is based on the researcher’s knowledge and experience of the group to be sampled using clear criteria to guide the process The potential for inaccuracy in the researcher’s criteria and resulting sample selection limits the ability of the researcher to generalize the results

Quota Sampling Selecting a sample based on required, exact numbers, or quotas, of individuals or groups of varying features Widely used in large-scale surveys when data are obtained from easily accessible individuals within well-defined categories People who are less accessible (e.g. more difficult to contact) are underrepresented.

Qualitative Sampling Much smaller sample – qualitative studies with more than 20 samples are very rare, Interaction between informant and researcher is longer in time and greater in depth. The participant’s opinions, thoughts, and feelings are crucial to a qualitative study.

Types of Qualitative Sampling Intensity Sampling – selecting participants who are at different levels. For example, a participant who is at the beginning of their college career, and one about to graduate. Homogeneous Sampling – selecting participants who are extremely similar in their experience. For example, two students in the same grade performing at the same academic level. Criterion Sampling – selecting participants who meet some part of the research criteria. For example, the participants meet some sort of predetermined set of criteria – does not have to be from the same level, or group.

Types of Qualitative Sampling Snowball Sampling– utilizing initial participants to recruit additional participants for study. For example, if you are conducting a study in which it is difficult to find participants, you may start with one or two participants and utilize their connections to other people for more participants. Random Purposive Sampling – selecting more participants than planned for the study, and then randomly selecting who from the group you will use to participate. For example, surveying 30 people for your study, but only using 15 for actual data within the study. *used primarily with very small samples.

Quantitative vs. Qualitative Sampling Which type of sample will you use for your research?