Research Bias, Problem and Theory

Please do the questions separately. 
Topic 2 DQ 1 
Discuss sources of bias for both quantitative and qualitative research. For quantitative research, be sure to address both random and systematic bias. You may use examples from the articles you selected as illustrations of bias and/or preventing bias.
Topic 2 DQ 2 
Researchers often identify the research problem and then go in search of a theory. Discuss the disadvantages of doing this. What does the textbook recommend that researchers do to assure a true fit between theory and designing the study?
LECTURE NOTES
Quantitative and Qualitative Research Designs
Introduction
In order to understand the research process, it is helpful to identify those elements that must be incorporated in any study design within the two paradigms: quantitative and qualitative. There are common elements in planning any type of research study. In the same way that there are some commonalities in providing care to acute care patients, there are also unique differences due to age, gender, general physical health, and type and severity of illness. Once these common elements are understood, characteristics of qualitative versus quantitative designs need to be differentiated. Philosophy Science and Nursing, will assist you in understanding the different types of qualitative and quantitative study designs and some of the assumptions about both types of research. Also a researcher must understand the notion of validity or trustworthiness which is the ultimate criterion for evaluating a study. For quantitative designs, there are a number of strategies that can be implemented to enhance rigor, and therefore validity, of the findings.
Elements Common to Planning a Study
All research studies are intended to generate new knowledge that is of some importance. Though the designs may vary, all have to address the truthfulness of the results.
The validity of a study refers to the extent to which findings are true. Internal validity is the extent to which the study findings are true and trustworthy. External validity refers to the extent to which the study results can be applied to people outside of the study.
Validity, whether internal or external, means that the measurement used is measuring what it is supposed to measure. For example, a thermometer measures temperature; for more abstract variables like pain, assuring validity is more complex. A valid measurement of pain for an adult (e.g., on a scale of 1 to 10) would not be valid for an infant. An instrument is considered reliable if it measures the same way each time. The degree of reliability of the selected measurement should be the same each time it is used. If consistency of measurement cannot be achieved a new instrument should be utilized. An electronic thermometer is a valid measurement of temperature, but if not calibrated correctly it may not be reliable.
No matter the type of study design, steps are taken to assure the validity of the results. While validity is the word used for quantitative research, trustworthiness is the equivalent word used when referring to results of qualitative studies. Both are concerned with the truth of the findings. The actual strategies used to ensure the truthfulness of research results vary according to the study design. Three elements common to all study designs are bias, timing, and setting.
Bias
Bias is a major threat to the internal validity of a study. For quantitative research, bias is minimized principally by strategies to exert control over any factors that might alter the relationship between the variables that are of central interest (Polit & Beck, 2012). For qualitative designs, bias is minimized by reflexivity. Reflexivity requires having researchers reflect upon their personal values and beliefs that might affect the accuracy of data collection and interpretation.
Timing
According to Polit & Beck (2012), the consideration of time is another element that is of importance for all designs. For qualitative researchers, there must be adequate time for the researcher to capture the fullness of the phenomenon of interest. For quantitative researchers, time is a critical element for both the study design and collection of data. All human experience is affected by time (e.g., time of day, length of time, past, present, future).
Setting
The setting for the research will have an impact on both internal and external validity. Human beings are strongly affected by their environments. Therefore, the location of the study must be thoughtfully selected and fully described so that consumers of the research can evaluate the applicability of the findings to their own situation and setting (Polit & Beck, 2012).
Enhancing Rigor in Quantitative Designs
Enhancing rigor in quantitative designs, whether experimental or nonexperimental, refers to strategies aimed at eliminating threats to the validity of findings such as minimizing biases and controlling variables. To further cloud the use of terminology, Polit and Beck (2012) refer to four types of validity: construct validity, internal validity, external validity, and statistical conclusion validity.
Construct Validity
Construct validity refers to the extent to which measurements of variables, and interventions, are accurate representations of what was intended. For example, when studying an educational intervention to change attitudes about AIDS, is it reasonable to think that a 10-minute video would be effective? The development of an effective educational intervention requires knowledge of the learning theory (e.g., cognitive, affective, psychomotor), characteristics of the learner, and the kinds of information required to alter someone's beliefs.
Internal and External Validity
In order to have assurance that the effect observed is real and not due to other causes, internal validity is enhanced by addressing the issues related to causality. In addition, since statistical significance is required to demonstrate that a relationship exists between the cause of study and its effects, statistical tests have properties that must be considered when designing data analysis. The appropriate size of the sample and level of measurement must meet the requirements of the statistical procedure for the test to be able to detect differences. However, it is also important for the presumed cause to be measured adequately. The major strategy used is to estimate the power of the test. Each design has particular strengths and limitations; the strategies to increase rigor vary according to the design (Polit & Peck, 2012).
External validity is affected by a number of factors: the setting; the characteristics of the sample; the characteristics of the intervention (e.g., would other individuals be as effective in the teacher role of a study); and the extent to which the research situation is similar to the real world. Unfortunately, the strategies used to decrease threats to internal validity tend to make the experimental situation less like the real world. What may work in a highly controlled environment may not have the same effect in a naturalistic setting.
Statistical conclusion validity should be the first tests administered in the study. The statistical validity confirms that there is a valid and empirical relationship between the presumed cause and the effect (Polit & Beck, 2012). It determines whether or not the test was administered fairly and identifies possible causes of misinterpretation.
Conclusion
The purpose of the research and research question guides the researcher in the selection of the appropriate study design. The first decision is whether the approach should be quantitative, qualitative, or both (i.e., triangulation. Within each approach, there are many designs). The design, like the approach, is selected according to the purpose of the research and the ability of the design to answer the specific research question, rather than the researcher's familiarity with the study design. The goal of the study is to discover the truth. All study designs have strengths and limitations. There are tradeoffs for selecting one design over another. It is the researcher's responsibility to implement strategies to improve the rigor of the study to assure findings that are valid and useful.
References
Polit, D. F., & Beck, C. T. (2012). Nursing research: Generating and assessing evidence for nursing practice (9th ed.). Philadelphia, PA: Lippincott, Williams & Wilkins.