Research Methods: Sampling Techniques PDF

Topic 6 Research Methodology (Part 2) Subtopic: D. Sampling Procedure (Non-Probability) E. Data-Gathering Procedure F. Research Instruments Objectives: At the end of the topic the students are expected to be able to: a) make logical decisions as to the appropriate data collection meth...

Topic 6 Research Methodology (Part 2) Subtopic: D. Sampling Procedure (Non-Probability) E. Data-Gathering Procedure F. Research Instruments Objectives: At the end of the topic the students are expected to be able to: a) make logical decisions as to the appropriate data collection method for specific studies b) design questionnaires/interview schedule for their research proposal; and c) Discuss the advantages and disadvantages of each data collection method. B. NON PROBABILITY SAMPLING Non-probability sampling – is a process whereby probabilities cannot be assigned objectively to individuals in the population. - the probability of selecting population elements is unknown. There are seven types of non-probability sampling, namely: 1. convenience sampling 5. voluntary sampling 2. purposive sampling 6.snowball sampling or networking 3. quota sampling, 7. accidental sampling. 4. dimension sampling  B. NON PROBABILITY SAMPLING cont.. 1. Convenience sampling– selecting sampling units that are easily (conveniently) available to the researcher. Ex. due to budget and time constraints, a researcher may just select students from his class to investigate the relationship between educational history and academic achievement of students. Or he may select the first 100 students who are available.  B. NON PROBABILITY SAMPLING cont… 2. Purposive/Judgment sampling – a process whereby you select a representative sampling according to your subjective judgment. selecting units to be observed on the basis of our judgment about which one will be useful or representative. Ex. If a researcher wants to study the value chain analysis of vegetables, he will purposely select those involved in vegetable production and marketing. Advantage and Disadvantage of Purposive/Judgment Sampling Advantage: it assures some information from respondents who are hard to locate and data crucial to the study can be obtained. Drawback: there is no assurance that those selected are in any way representative of the target population  B. NON-PROBABILITY SAMPLING cont… 3. Quota sampling – this is simply a variation of judgment sampling which provide more explicit instructions on what to select - selecting samples on the basis of pre-specified characteristics, so that the total sample will have the same distribution of characteristics as assumed to exist in the population being studied. Advantage & Disadvantage of Quota sampling Advantage: it does not require the population lists or dwelling-unit maps that are needed for probability sampling. Disadvantage: there are some doubts about its representatives in relation to the population. B. NON PROBABILITY SAMPLING cont… 4. Dimensional Sampling (according to Cooper and Emory, 1995)- a multi-dimensional extension of quota sampling. In this sampling procedure, instead of a large size, a small size is selected, It is emphasized that all areas of interest should cover at least one case.  Dimensional sampling A dimensional sample is a type of sampling technique used in research where participants are selected based on specific, predefined dimensions or characteristics that are relevant to the study. These dimensions might include variables like age, income, education level, gender, occupation, or geographic location. The aim is to ensure that all important dimensions are represented in the sample, allowing researchers to make comparisons across different groups. Example of a Dimensional Sample: Suppose a researcher is conducting a study on consumer preferences for electric vehicles (EVs) in a city. The researcher wants to ensure that the sample reflects differences in preferences across certain dimensions, such as income level and age group. Dimensions to be considered: Income levels: Low-income, middle-income, and high-income groups. Age groups: 18-30, 31-50, and 51-70 years old. 5. Voluntary Sampling 5. Voluntary Sampling – a special type of sampling in which subjects/cases are informed about the subject matter willingly or voluntarily participate in the study. ❖This sampling is useful especially if one is dealing with information on sensitive or delicate issues. B. NON PROBABILITY SAMPLING cont… 6. Snowball Sampling (networking sampling – a type of sampling in which a researcher first identifies few individuals for the sample and uses them as informants. - useful when one wants to consider possible respondents who are not normally visible. ( Drug Addicts, Patients with Aids, etc.) 7. Accidental sampling- a type of sampling method wherein who happens to be in the area at the time of the incident form part of the sample. Types of Sampling T h e different types of s am pling c a n b e s u m m a r i z e d in Figure 11. Simple R a n d o m Lottery Technique Sampling U s e of T a b l e of R a n d o m N u m b e r s Systematic R a n d o m Probability Sampling Proportionate Stratified S a m p l i n g Sampling Stratified R a n d o m Sampling Disproportionate Stratified S a m p l i n g Cluster Sam pling Multi-S t a g e S a m p l i n g Strip Sam pling T y p e s of Sam pling Convenience Sam pling Purposive Sampling Q u o t a Sampling N o n Dim ension Sam pling Probability Sam pling Voluntary Sam pling Snowball Sampling/ Networking Accidental Sam pling Fig. 1 1 T yp e s of Sam pling DETERMINATION OF A SAMPLE SIZE How large must a sample be? Sample size depends upon the following (McMillan,1992) a) researcher’s finance b) problem c) categories to be employed in the analysis of data d) nature of the population (whether it is homogenous or heterogeneous), and e) ultimate purpose for which the study is undertaken (whether for theoretical or for practical ends). DETERMINATION OF A SAMPLE SIZE cont.. Slovin, as cited by Sevilla et.al. (1992), for instance, offers the following formula to estimate sample size: n= N 1 + Ne2 Where: n= sample size desired; N= population size; and e= desired margin or sampling error DETERMINATION OF A SAMPLE SIZE cont.. Using this formula, what would be the sample size if the total population (N) is 2,000 and the margin of sampling error you allow is 5%. The sample size may be computed this way: n = 2000 1 + (2000) (.05)2 n = 2000 1+5 n = 2000 6 n= 333 Sample Size Table 5. Sample size for specified margins of error based on Slovin’s formula. Margins of error Populati +1% +2% +3% +4% +5% +10% on 500 * * * * 222 83 1,500 * * 638 441 316 94 2,500 * 1,250 769 500 345 96 3,000 * 1364 811 517 353 97 4,000 * 1,538 870 541 364 98 5,000 * 1,667 909 556 370 98 6,000 * 1,765 938 566 373 99 7,000 * 1,842 959 674 378 99 8,000 * 1,905 976 580 381 99 9,000 * 1,957 989 584 383 99 10,000 5,000 2,000 1,000 588 385 99 50,000 8,333 2,381 1,087 617 387 100 Minimum Acceptable Sample Size Gay, as cited by Sevilla et. al. (1992) offers some minimum acceptable sizes depending on the type of research, without necessarily following a formula. a. Descriptive research – 10 percent of the population, for small populations, a minimum of 20 percent may be required. b. Correlation research – 30 subjects c. Ex post facto or causal – comparative research – 15 subjects per group d. Experimental research – 15 subjects per group (other authorities believe that 10 per group should be considered minimum). E. DATA COLLECTION PROCEDURE Data collection is any process of preparing and collecting data. Purpose of data collection: 1. obtain information to keep on record, 2. make decisions about important issues 3. Pass information on to others.  SOURCES OF DATA Sources of data are generally categorized into primary and secondary sources. 1. Primary sources of data are those that provide information that are collected for the first time as part of a research project. 1. Secondary sources are those that provide data which have been collected previously and reported by some individual other than the present researcher. ETHICS IN RESEARCH The following are some key ethical principles or standards that should underlie any research endeavor (Trochim, 2001; Best and Kahn, 1998, APA, 1992): 1. The Principle of Voluntary Participation. This principle requires that people should not be coerced into participating in research. 2. The Principle of Informed Consent. This principle requires that prospective research participants must be fully informed about the procedures and risks involved in research and must give their consent to participate. ETHICS IN RESEARCH cont… 3. The Principle of No Risk of Harm. Participants of the research study should not be put in a situation where they might be at risk of harm (whether physical or psychological) as a result of their participation. 4. The Principle of Privacy. This principle requires the protection of the privacy of research participants. There are two sub-principles under this, namely: The Principle of Confidentiality. Participants of the research study should be assured that identifying information is not be made available to anyone who is not directly involved in the study. ETHICS IN RESEARCH cont… b. The Principle of Anonymity. This principle requires that participants will remain anonymous throughout the study even to the researchers themselves. 5. The principle of Equality of Service. - the person’s right to service. - happens in experimental studies where a no-treatment control group is assigned. -subjects assigned to the control group may feel that the rights to equal access to services are being curtailed. ETHICS IN RESEARCH cont… 6. The Principle of No Deception. According to the American Psychological Association (1992), participants should not be deceived about the purpose or nature of the research unless justified by the study’s prospective scientific, educational, or applied value and that equally effective alternative procedures that do not use deception are not feasible. 7. The Principle of Knowledge of Outcome. Research participants have the right to receive an explanation of the results of the investigation. - Explanation may be done orally, in writing, or by informing them of any publication in which the research report is published. F. RESEARCH INSTRUMENT Research instrument - is what you use to collect the information in a qualitative field study or observation. It helps you keep track of what you observe and how to report it. It must be both valid and precise. Evaluation of an Instrument Cortes (1994) suggests these questions to help the researcher ascertain whether the instrument he is using is “good” or not: Is the tool appropriate for the study? Was there a trial run of the tool to determine the difficulty and validity indices of the items included? Are the items in the instrument relevant to the problem on hand? How long does it take to finish answering the instrument? Are the questions clearly stated? Has the instrument stood the test of time? How popular is it? What are the critiques on its use? Were these considered? Will responses yield to quantification and descriptive qualification? Is the instrument easy to administer? Is scoring facilitated.  METHODS OF COLLECTING PRIMARY DATA Questionnaires * case-studies Interviews * diaries focus group interviews * critical incidents and Observation * portfolios. 1. QUESTIONNAIRE 1. Questionnaire- a set of printed or written questions with a choice of answers, devised for the purposes of a survey or statistical study. - a data collection too which properly sets out the way in which the research questions of interest should be asked to produce precise information. 3 main parts of a questionnaire 1. Introduction it must be persuasive it must qualify the respondent as someone who belongs to the sample 2. Body or content - It consist of questions that cover information needed to solve the problem 3. Basic data – covers mainly information about the respondent  Advantages & Disadvantages of different questionnaires Mode of Data Advantages Disadvantage Collection Personally Can establish rapport and Explanation may administered motivate respondents introduce a bias Questionnaires Doubts can be clarified Take time and effort. Less expensive when administered to groups of respondents. Almost 100% response rate ensured Anonymity of respondent is high  Advantages & Disadvantages of different questionnaires Mode of Data Advantages Disadvantage Collection Mail Anonymity is high. Response rate is almost questionnaires Wide geographic always low. A 30% rate regions can be is quite acceptable. reached. Cannot clarify questions Respondents can take Follow-up procedures more time to respond for nonresponses are at convenience. necessary. Can be administered electronically if desired.  Advantages & Disadvantages of different questionnaires Mode of Data Advantages Disadvantage Collection Electronic Easy to administer Computer literacy is a questionnaires Can reach globally. must. Very inexpensive Respondents must have Fast delivery access to the facility. Respondents can Respondents must be answer at their own willing to complete the convenience like the survey.. mail questionnaire. Less pressure on the part of the respondents  2. Interview Interviewing - a technique that is primarily used to gain an understanding of the underlying reasons and motivation for people’s attitudes, preferences or behavior. Interview can be undertaken an a personal one-on- one basis or in a group. - can be conducted at work, at home, in a street or in a shopping center, or some other agreed location.  2. Interview cont.. Types of Interview A. personal interview – face to face contact between the interviewer and respondents. B. telephone interview- a data collection method by which there is only voice contact between the interviewer and the respondent. 2. Interview cont.. Three basic types of interviews (Berg, 1995). 1. Scheduled-structured interview – uses an instrument in which the questions, their wording, and their sequence are fixed and are identical for very respondent. 2. Nonscheduled-structured interview – uses only guide questions for the interview 3. Nonscheduled interview – does not use pre-specified set of questions. The interviewee does most of the talking, with little or no direction from the interviewer. 2. Interview cont.. B. Advantage of telephone interview: Less expensive with relatively rapid completion and high response rate Limitations of telephone interview researcher is limited to telephone subscribers which generally are not representative of the population quite impossible to conduct a lengthy interview over the telephone. Thank you for listening!