Sampling and Data Collection in Research
Judith A. Geske
University of Phoenix
Sampling and Data Collection in Research
Sampling and Data Collection in Research are tools that assist researchers to gather important information regarding a specific group of people (Monette, Sullivan, & DeJong, 2011). There are fundamental steps that need be followed to practice accurate research as much as possible. For example, probability and nonprobability as stated by Monette, Sullivan, & DeJong, 2011, “Researchers use probability samples in some types of human service research; such as needs assessment and evaluation research. Nonprobability samples are used in assessing client functioning and evaluating the effectiveness of intervention strategies.” In addition to sampling, data collection is as essential as sampling. Reliability and Validity are two important scales of measure examples of research used in human service management (Monette, Sullivan, & DeJong, 2011). It is critical to know the advantages and disadvantages in data collection methods. Some of the different methods used are: telephone surveys, online surveys, focus groups, and surveys via websites. Sampling
The purpose of sampling is to study a part of a whole group been studied. According to Monette, Sullivan, & DeJong, 2011, some groups are just two big and sampling allows the study of a workable number of cases from the large group to derive findings that are relevant to all members of the group (Chapter 6, The Purpose of Sampling). One example of probability sample is Simple Random Sampling (SRS), such as trying to do a research project that calls for a national sample of 2,000 households. SRS is considered the basic sampling procedure on which statistical theory is based. (Monette, Sullivan, & DeJong, 2011). An example of nonprobability sample is Ronald Feldman and Timothy Caplinger (1977) that were interested in factors that bring about behavior changes in young boys who exhibit highly visible antisocial behavior. This example requires students to meet periodically and a commitment of time and energy. This method required considerable resources which lacked availability (Monette, Sullivan, & DeJong, 2011). Research in human services should avoid bias when selecting samples. According to Monette, Sullivan, & DeJong, 2011, “systematic sampling can produce biased samples” (Chapter 6, The Purpose of Sampling). The problem occurs when a cyclical pattern arises because if an interval occurs than it is possible to get a biased sample (Monette, Sullivan, & DeJong, 2011). Caution needs to be taken into consideration when a human service agency is doing a research on certain group so that biased errors do not occur. Data Collection
Scales of measurement as explained by Monette, Sullivan, & DeJong (2011) are indicators, item, and index that serve as tools to measure variables (Chapter 5, Applied Social Research). Indicators are observations that are assumed evidence of the attribute or properties of some phenomenon (Monette, Sullivan, & DeJong, 2011). Item is a single indicator of a variable that can take the form of an answer to a question (Monette, Sullivan, & DeJong, 2011). Next variable is index, it refers to a combination of a number of items into a composite score (Monette, Sullivan, & DeJong, 2011). Among data collection methods reliability is one that refers to a measure’s ability to yield consistent results each time it is applies (Monette, Sullivan, & DeJong, 2011). Some examples of reliability methods are Test-Retest, Multiple Forms, and Internal Consistency Approach. Test-Retest is a method that refers to applying the same technique to the same group. Multiple Forms refers to the technique of multiple forms with separate but equivalent versions made up of different items; and Internal Consistency method approach uses a single scale that is administered to one group of people to develop an estimate of reliability (Monette, Sullivan, & DeJong, 2011). Reliability methods are important to measuring of data collection in the human service profession because, they make information reliable. The systems used to apply these methods are consistent in its results as stated by Monette, Sullivan, & DeJong, 2011. In addition to reliability there is validity measures that refer to accuracy of measure (Monette, Sullivan, & DeJong, 2011). There are different methods of validity, among the ones included on the text book are: Face validity, criterion validity, current validity; predictive validity, and construct validity (Chapter 5, Applied Social Research). Each of these examples of data measurement demonstrate valid measurement of data collection. Validity is not as easy to demonstrate with any finality (Monette, Sullivan, & DeJong, 2011). It is important to make sure that data collection method and instruments are both reliable and valid because in the field of human services the lives of people depend on the findings of these methods. The help and methods used cannot be bias but as accurate and efficient as possible. Advantages and disadvantages of the following methods of data collection are as follow: 1.Telephone surveys are not as expensive as face-to-face and can be done at a faster speed; can be used in a multilingual way; and supervision is much easier. Some of the disadvantages of telephone interviews are the limitations of questions been asked; there is no personal contact only voice to voice; lack of effectiveness in the inability to present complex sets of questions that are important (Monette, Sullivan, & DeJong, 2011). 2. Online surveys as stated by Fricker and Schonlau 2002; Schonlau et al. 2004), there are many benefits to online or email surveys. Among the benefits are speed, low cost, and ability to reach respondents anywhere in the world; are less expensive and the responses are returned much more quickly. This method of interview has its disadvantages are some of those are: Not everyone has access to the internet; not everyone choses to respond to requests to fill out an online survey; and the population of people who have access to internet tend to be affluent, well educated, young and male (Duffy et al. 2005; Kaplowitz, Hadlock, and Levine 2004; Schonlau et al. 2004). 3. Support Groups are flexible, cost less, and provide quick results among other advantages. Disadvantages of support groups are less generalizable, and data is more difficult to analyze (Monette, Sullivan, & DeJong, 2011). The method that I would use would be support groups. I think is more personal and more engaging. In summary, the methods used for sampling and data collection are all ways to study a certain group with a problem. As a human service worker researching about a certain group, helps in better helping the client. The more a human service knows about a problem the more efficient the help can be.
Monete, D. R., Sullivan, T.J., & Dejong, C. (2011). APPLIED SOCIAL RESEARCH (8th ed.). Retrieved from The University of Phoenix eBook Collection database.