Applying the Principles of Quantitative Data Collection
Introduction
Every researcher has a responsibility to protect the participants in an investigation. As part of their investigation researchers ask specific questions then use their intuition, curiosity, training, and statistics to answer the research question. Good scientists do not simply make decisions; instead, they use tools to determine the best course of action to make informed decisions. This paper will identify a research article, use a decision tree to map three decision points that informed the data collection process, and focus on the feasibility and ethical dimension of the study.
Data Collection Process
Khatiban and Sangestani’s (2014) study was chosen for this review to evaluate the effects of problem-based learning [PBL] as compared with the traditional method, non-problem-based learning [NPBL], in students’ clinical nursing education. The sample represented a random sampling of 70 third-year nursing students assigned into two groups, as either PBL or control group, in Hamadan University of Medical Sciences in Iran. The study focused on three research questions. First, will the students in the comparison group showed higher levels of knowledge and skill compared to the students in the control group that did not participate in the intervention? Second, will the students in the comparison group showed higher levels of using the nursing process of assessment, diagnosis, planning, implementation, and evaluation compared to the students in the control group that used science to solve clinical problems? Third, will the students in the comparison group showed more positive attitudes toward their learning experiences in using the nursing process compared to students that did not use the nursing process?
To answer the research questions Khatiban and Sangestani (2014) selected a quasi-experimental research design. The research instruments used in the Khatiban and Sangestani (2014) study were three questionnaires and a checklist. Two tests were used to measure the variables; independent sample t-tests for the between-group comparison of competency self-evaluation, and paired sample t-tests for the between-group comparison of competency self-evaluation.
Decision Tree
Based on the decision tree (See Appendix A) Khatiban and Sangestani (2014) were interested in answering the question, would using a problem-based program in a clinical course make a difference between the PBL and NBPL group of students? One manner in which Khatiban and Sangestani (2014) sought to answer the question was by investigating the level of knowledge, skill, the nursing process, and attitude between two groups of nursing students that were independent of one another. The research questions asked were, (1) will the students in the PBL group show higher levels of knowledge and skill compared to the students in NPBL group that did not participate in the intervention, (2) will the students in the comparison group show higher levels of using the nursing process of assessment, diagnosis, planning, implementation, and evaluation compared to the students in NPBL that used science to solve clinical problems, and (3) will the students in the comparison group show more positive attitudes toward their learning experiences in using the nursing process compared to the students that did not use the nursing process?
The logical decision after considering the research questions and variables lead Khatiban and Sangestani (2014) to decide on using a quasi-experimental design with a nonequivalent control group pre and posttest design, and a nonequivalent control group only posttest design for students’ attitude and performance. The next step in the process was deciding what test to use to measure the variables. There were two t-tests used for the study which were an independent samples t-test for the between-group comparison of competency self-evaluation and a paired samples t-test for the between-group comparison of competency self-evaluation. Khatiban and Sangestani (2014) compared the means of two groups using a dependent samples t-test. A dependent samples t-test, also called a paired-samples t-test or correlated-samples t-test, is used to compare the means of two groups when individual scores in one group are paired or correlated with particular scores in other groups (Salkind, 2014). A t-test for the dependent indicates that a single group of the same subjects are being studied under two conditions (Salkind, 2014). In the Khatiban and Sangestani (2014) study, the PBL group of students was tested twice, before the start of the PBL program and after the conclusion of the clinical class.
Implication
In a quasi-experimental study, Khatiban and Sangestani (2014) evaluated the effects of PBL as compared with the traditional method, NPBL, in students’ clinical nursing education. Khatiban and Sangestani (2014) used two different tests for two different variables, independent samples t-test for the between-group comparison of competency self-evaluation and paired samples t-test for the between-group comparison of competency self-evaluation. Khatiban and Sangestani (2014) also applied a t-test for independent means. The independent-samples t-test, commonly referred to as a between-groups design, evaluated the differences between the means of the PBL and control groups of students to find out if the scores of competency self-evaluations for the two independent groups were significantly different from each other. One of the first steps in using the independent samples t-test is to test the assumption of independence. The assumption of independence means that scores of one participant are not systematically related to scores of the other participants. If the independence of assumption is violated it can ruin a study. The independence of assumption was assessed by Khatiban and Sangestani (2014) at the beginning of the study by confirming the two groups were independent of each other. If the independence of assumption was not dealt with or was violated it would not be feasible to use the study’s sample data to test the validity of the prerequisite condition. Any breach of integrity during the development, execution, or dissemination of results, whether it be intentional or unintentional, will seriously weaken or even invalidate a research study (Drew, Hardman, & Hosp, 2008).
A second point on the decision tree in the Khatiban and Sangestani (2014) was to compare the means of two groups using a dependent samples t-test. A dependent samples t-test, also called a paired-samples t-test or a correlated-samples t-test, is used to compare the means of two groups when individual scores in one group are paired or correlated with particular scores in other groups (Salkind, 2014). A t-test for dependent means is used to test sample means in order to determine whether there was a statistically significant difference in the mean of the dependent variable between two related groups (Salkind, 2014). In the Khatiban and Sangestani (2014) study the PBL group of students was tested twice, before the start of the PBL program for a baseline of the student’s knowledge before treatment was applied, and again after completing the clinical class. The data collected from the students are linked and then computed to find out if there was a difference between pairs or pretest and posttest scores. Random sampling is required for all statistical inference because it is based on probability (Polit & Beck, 2004).
Even though subjects were reported to be assigned to treatment at random, there may be some concern that any difference in the posttest measurements could be due to a failure in the randomization. If indeed the posttest measurements were due to a failure in the randomization, how would this failure affect the empirical or ethical soundness of the result of the study? Could this produce false results about PBL intervention in a clinical program? Ethical issues in quantitative research focus on protecting individuals who receive an intervention (Drew et al., 2008). One reason for this is that participants are encouraged to discuss their feelings, attitudes, and experiences, some of which may be personal. Therefore, the researcher must fully ensure that participants know that they can withdraw from a study and to discontinue the activities at any time. If the consent process is handled appropriately and participants know and understand their rights, problems that arise should be lessened and perhaps eliminated. The most basic concern in all research is that no individual is harmed by serving as a participant in a study (Drew et al., 2008).
A third point on in the Khatiban and Sangestani (2014) study was the grouping variable. The independent variable is sometimes called the grouping variable because each group has a specific level or value of the variable. All members of each group will receive or participate in the same intervention however it will be different for different groups. Grouping variables was helpful in the Khatiban and Sangestani (2014) study because two groups are compared which are the levels of the independent variable. The two groups were compared on the independent variable PBL so that any differences between the groups can be attributed to PBL in the clinical program. In the Khatiban and Sangestani (2014) study a concern was that the anonymity and privacy of the study participants were not disclosed, nor how anonymity and privacy would be maintained throughout the study including data gathering, data analysis, and data interpretation. Biases were not identified or how any would be addressed. It is not known what the risk and benefits were to the participants who volunteered for the study. The issues of anonymity and privacy are ethical issues that could be harmful to participants (Sieber, 2001) and should have been addressed in the study since this information would alleviate any speculation students may have about participating in a study.
References
Drew, C. J., Hardman, M. L., & Hosp, J. L. (2008). Measures and instruments. Designing and conducting research in education. Thousand Oaks, CA: SAGE Publications, Inc.
Khatiban, M., & Sangestani, G. (2014). The effects of using problem-based learning in the clinical nursing education on the students’ outcomes in Iran: A quasi-experimental study. Nurse Education in Practice, 14(6), 698-703. doi:10.1016/j.nepr.2014.10.002
Polit, D. F., & Beck, C. T. (2004). Nursing research: Principles and methods (7th ed.). Philadelphia, PA: Lippincott Williams & Wilkins.
Salkind, N. J. (2014). Statistics for people who (think they) hate statistics (5th ed.). Thousand Oaks, CA: SAGE Publications.
Sieber, J., E. (2001). Summary of human subjects protection issues related to large sample surveys. Retrieved from http://www.bjs.gov/content/pub/pdf/shspirls.pdf.
Appendix A
Decision Tree
| RQ 1, 2 & 3 want to know if students in a comparison group will show higher levels of knowledge and skill, using the NP and a more positive attitude than students in the control group. |
| Research Questions
RQ1: Will the students in the PBL group show higher levels of knowledge and skill compared to the students in control group that did not participate in the intervention? RQ2: Will the students in the PBL group show higher levels of using the nursing process of assessment, diagnosis, planning, implementation, and evaluation compared to the students in control that used science to solve clinical problems? RQ3: Will the students in the PBL group show more positive attitudes toward their learning experiences in using the nursing process compared to the students that did not use the nursing process?
|
| Differences between/among groups |
| Type of design
Quasi-experimental Quasi-experimental design measures the dependent variable (knowledge and skill, the nursing process, and attitude) |
| Variables = 2 types
Independent: Problem Based Learning Dependent: knowledge and skill, the nursing process, and attitude. |
| What test measures differences between two groups, i.e., the PBL and control group? |
| Grouping of Variable Equals Independent Variables
Nominal |
| t-test
Independent sample samples t-test for the between-group comparison of competency self-evaluation, paired samples t-test for a within-group test as is used to compare the pre and post test scores for only one group. |