Module 3 DQ 1 and 2 - Assignment Example

Statistics Number: Question Despite the documented merits that experimental and control group designs enjoy, there are other alternative designs that are used frequently. Survey research designs and cohort research designs are some of the alternatives that may be used. In the former, there is the collection of data, either through personal interviews or questionnaires, with the interviewee belonging to a designated population. As for the cohort research design, a study is to be conducted over a timeframe, and it has to involve a member of a population as a representative or subject member.

The representative member must have some elements of commonality uniting him with the population. There are several reasons why survey and cohort research designs are used. When the need to determine general trends, public perception and opinion towards the government, its policies, businesses, healthcare institutions or sector and the media arises, survey research designs comes in handy. The same can be used in both small-scale and large-scale levels of research studies. After determining the aim of the research design, the sample group is determined, in order to ensure accuracy.

As a form of research design, surveys are flexible enough to be carried out either on a face-to-face basis, or via mail (Turlik, 2010). Conversely, cohort design can be used in both medical science and social science researches. This type of experimental design is very popular because, it factors statistical occurrences within a designated subgroup, and hence upholding the highest extent of accuracy possible. The results accrued are valid since the subgroup has similar characteristics that are pertinent to the problem being researched on.

The cohort design is popular because of its flexible nature, its compatibility with both primary and secondary data and its affiliation with ethical standards. Question 2Internal Validity in a Research Study Internal validity has to do with: how well a research study has been carried out, in light of operational definitions and the research design used and the measuring of the variables; and the extent of confidentiality that is adduced when observing the effects brought about by independent variables, in lieu of extraneous variables.

Again, internal validity covers accuracy of the results produced. If for instance sampling is done inchoately, results obtained may be inaccurate. Impacts of Several Errors or Biases That Can Reduce the Studys Validity Errors and biases can undermine the validity of a research study in several ways. First, the failure to do random sampling will automatically generate inaccurate results. For instance, if a designated sample only comprises Hispanic patients only, then attributing the results obtained will be incorrect, if the results are generalized to serve the entire American population (Jiménez-Buedo and Miller, 2010).

In another wavelength, bias mainly exhibits itself in research study as selection bias. In selection bias, differences among groups are extant even at the pre-test stage, and can interact with independent variables, and thereby relaying a given outcome. When the selection being done factors an unequal number of samples with variables that are similar and subject-related, then the findings will also not be accurate. Because of this bias, the study will be bereft of internal validity. The same applies to the error of instrument change.

Internal validity is compromised when the researcher sidesteps the importance of retrospective pre-tests and proceeds to alter the criteria being used to make conclusions or judgments. ReferencesJiménez-Buedo, M. & Miller, L. M. (2010). “Why a Trade-Off? The Relationship between the External and Internal Validity of Experiments.” Theoria, 25 (3), 301-321. Turlik, M. (2010). “Evaluating the Internal Validity of a Randomized Controlled Trial.” Podiatry Management, 29 (3), 191-4.