FFR-J5

RCH 5302, Foundation for Research 1

Course Learning Outcomes for Unit V Upon completion of this unit, students should be able to:

4. Explore the dynamics of data sampling and distributions. 4.1 Examine the use of data sampling and distributions in a data analysis plan.

5. Analyze research data through the application of statistics.

5.1 Explain strategy and choices of analysis in a research study.

Course/Unit Learning Outcomes

Learning Activity

4.1

Unit Lesson Chapter 12, pp. 361–368 Appendix G, pp. 472–474 Article: “Perceived Managerial and Leadership Effectiveness in a Korean

Context: An Indigenous Qualitative Study” Article: “Opportunity and Rationality as an Explanation for Suspicious Vehicle

Fires: Demonstrating the Relevance of Time, Place, and Economic Factors”

eBook: Fundamentals of Research Methodology : A Holistic Guide for Research Completion, Management, Validation and Ethics, pp. 75–86, 89–104

Unit V Project

5.1

Unit Lesson Chapter 12, pp. 361–368 Appendix G, pp. 472–474 eBook: Fundamentals of Research Methodology : A Holistic Guide for

Research Completion, Management, Validation and Ethics, pp. 75–86, 89–104

Unit V Project

Required Unit Resources Chapter 12: Organizing Data and Analyzing Results, pp. 361–368 Appendix G: Scale Types and Associated Statistical Analyses for Common Research Approaches, pp. 472– 474 In order to access the following resources, click the links below. Read Section B: Data Collection and Analysis on pp. 75–86 and 89–104 from the eBook below. Godwill, E. A. (2015). Fundamentals of research methodology : A holistic guide for research completion,

management, validation and ethics [E-reader version]. https://libraryresources.columbiasouthern.edu/login?url=http://search.ebscohost.com/login.aspx?direc t=true&db=nlebk&AN=1023403&site=eds-live&scope=site&ebv=EB&ppid=pp_75

Read pp. 795–809 of the following article for an example of sampling, data collection, and data analysis in a qualitative study.

UNIT V STUDY GUIDE

Data Collection, Data Analysis, and Statistics

RCH 5302, Foundation for Research 2

UNIT x STUDY GUIDE

Title

Chai, D., Jeong, S., Kim, J., Kim, S., & Hamlin, R. (2016). Perceived managerial and leadership effectiveness in a Korean context: An indigenous qualitative study. Asia Pacific Journal of Management, 33(3), 789–820. https://libraryresources.columbiasouthern.edu/login?url=http://search.ebscohost.com/login.aspx?direc t=true&db=bsu&AN=117574075&site=ehost-live&scope=site

Read pp. 3–8 of the following article for an example of data sampling, collection, and analysis in a quantitative study. Kelly, H., Clare, J., Wuschke, K., & Garis, L. (2019). Opportunity and rationality as an explanation for

suspicious vehicle fires: Demonstrating the relevance of time, place, and economic factors. Crime Science, 8(1).

Unit Lesson An insightful point about data collection and the measurement that relates to the analysis is that any measure you use will not perfectly model reality. This “reality check” is much to the dismay of researchers. The analysis a researcher selects as a part of the research design may not completely measure what is needed to answer the research question or may inadvertently factor in variables or figures that are not intended or relevant to the research study. This inevitable truth about human efforts is why supervisors or the institutional review board may be rather exacting in their scrutiny of a researcher’s proposal. Indeed, after receiving presentation feedback on the data collection, the researcher may have to start over or add significant changes to the proposal in order to gain approval. This dynamic of power over researchers can cause potential frustration. Whatever the specific situation may be, it is also true that thorough preparation and attention to detail increases the chance that some of the research proposal or the subsequent study will not have to be redone or deemed a waste.

Reviewing Data Collection Quantitative and qualitative studies each have their own peculiarities regarding data collection. Quantitative data collection may inherently be labeled and coded already depending on the nature of the study, but in any case, labeling and coding data are virtues that are practically required. Given the complexities of life and nature, it is difficult to imagine a study in which collected data did not have to be labeled or coded in some way. Coding changes in variables, interviewee names, personal details, and responses allows for statistical analysis to be conducted, allows for big-picture conclusions to be made, and makes reporting results on charts or matrices more feasible. Certain applications, such as spreadsheets, are useful in coding, labeling, and charting data as it is collected. Coding prevents data from being accumulated in notebooks or other records, only to have the researcher forget what data goes where or what something meant. Mistakes in translating or coding will start to skew the findings and possibly the conclusions as well.

Data Analysis in Quantitative Studies In quantitative studies, what analysis do you choose? Your research problem and question(s) will guide you in this choice—as will literature from your review that will describe what has been tried before. You may be supporting business decisions with answering a research question; if so, a break-even analysis will likely be your analysis tool, and you may build a chart of these break-even points (BEPs), resulting from changing the variables in certain ways.

RCH 5302, Foundation for Research 3

UNIT x STUDY GUIDE

Title

Render, Stair, Hanna, and Hale (2015) show that if a researcher is looking for changes in a mean or expected value, they can use those common equations for data collection and analysis such as the one shown below.

A researcher who is analyzing with distribution calculations may select the exponential distribution method for variables of random values (such as time to completion) or the Poisson distribution method for discrete events (such as the number of arrivals per day in a department of motor vehicles office). Distributions, a common analysis tool, demonstrate that a dependent variable occurs most often at the calculated position (i.e., the mean or simple average) and that the probability that the variable has other values drops off to a certain calculated confidence level, upon which readers may be confident that the mean value is the likely result, proving or disproving the hypothesis.

Data Analysis in Qualitative Studies In qualitative studies, coding data as the information is being collected assists in analysis because descriptions can be hard to track as the amount of data accumulates in terms of writing or recordings. Coding in these studies helps protect anonymity and mitigate unintended biases in interpreting data for findings. Analysis can consist of descriptions of interview answers and their significance or what they indicate. Matrices can help show the trends of responses. A researcher in a qualitative study, after becoming familiar with the data by reading and reviewing, seeks to find themes among the trends of what has been said and categorizes data in themes or other ways that can be defended during the presentation. Patterns can be identified after some categorization; by describing these patterns and their possible significance, findings can be interpreted. Qualitative analysis takes some writing to accomplish this, and as with any good scholarly paper, the presentation in writing needs to be purposeful, sensible, concise, thorough in content, and effective in persuading all but the most intractable of readers that the study has resulted in the findings offered by the researcher. While there is no single standard way in which to format the various parts of a research report, it is helpful to start with a basic framework. A framework for the analysis section of a qualitative study using interviews could be organized and presented as shown below.

RCH 5302, Foundation for Research 4

UNIT x STUDY GUIDE

Title

• Methodology: Identify the methods employed to conduct the research. In this case, it is interviews, but what kind (i.e., personal, telephonic, or electronic)? Will primary or secondary interview data be used?

• Demographics: Identify background variables that describe the participants.

• Strategy behind the interview design: Explain the types of questions used and how they relate to the research problem or thesis.

• Data analysis: Explain the processing or organizing procedures used. (Cite by name when using specific tools or methods named for scholars who pioneered or first described them.)

• Interpretation of the data and conclusions drawn: The interview data (participant experiences) can be analyzed and organized into themes. Identify and describe each of the themes that emerged from the data analysis.

• Data processing: Once the raw data is analyzed, it must be examined to detect errors and omissions and to correct them. This ensures the data collected is accurate, consistent, and as complete as possible to better facilitate coding and tabulation.

In qualitative studies, writing a persuasive and detailed description in all of the above areas is a proven method for successful data analysis. Readers of the research report are led from presentations of themes that show where issues were discovered or reaffirmed to themes that indicate what the sampled population seems to be doing about it. They can then understand the meaning of the themes through the data organization and presentation and the descriptive interpretations and conclusions.

Data Analysis Tools Devlin (2021) has some useful features for planning data analysis. Appendix A on p. 449 of the textbook provides a decision tree and explanation to guide a researcher on how to analyze the data for a particular study. Note that analysis of the whole sample usually entails showing where the values seem to cluster—by measuring the mean, median, or mode—and showing how strong that tendency was by calculating the distribution probability. Questions about group differences, on the right of the decision tree, can be analyzed with statistics by running one of several tests depending on whether the data collection achieved possible causality or, more commonly in the natural world, only correlation. Appendix E, on pp. 457-464 of the textbook, offers more resources to measure, including those for qualitative studies. The commonly practiced way to use a scale is to select one and defend it in the methodology section by citing from sources and concluding that such a gathered coalition of literature demonstrates why the chosen analysis strategy is the best fit for the particular study. Appendix F, on pp. 466–471 of the textbook, and Appendix G, on pp. 472–474 of the textbook, support the decision tree in Appendix A with more detailed descriptions of how each measurement and statistics tool fits certain cases, such as how many dimensions of data are collected, how many dependent variables exist, and what type of measurement (e.g., ordinal or nominal) will be used.

Data Errors Contingencies to keep aware of as the proposal is being finalized are the possibilities of Type I and Type II errors, as explained by Devlin (2021) in Chapter 3 on pp. 89–98. A Type I error is a conclusion, usually confirming the hypothesis, when the data does not indicate that the hypothesis should be accepted. This acceptance of the hypothesis also is a rejection of the null hypothesis since the null hypothesis should have been accepted as true. A Type II error is failing to find the changes that should have led to acceptance of the hypothesis and rejection of the null hypothesis; instead, by Type II error, the null hypothesis would be erroneously accepted. Strategies to avoid Type I and Type II errors and to strengthen validity should be integrated in the research design and explained in the proposal and final report (usually in the methodology section). These are selected to fit the specific study and can consist of running significance tests to a certain level, raising the sample size to achieve acceptable randomness, or adding special safeguards against bias. As is often the case, literature searches on what has been done before provides the best authority on what to add to a research design.

References Devlin, A. S. (2021). The research experience: Planning, conducting, and reporting research (2nd ed.).

SAGE. https://online.vitalsource.com/#/books/9781544377933

RCH 5302, Foundation for Research 5

UNIT x STUDY GUIDE

Title

Render, B., Stair, R. M., Jr., Hanna, M. E., & Hale, T. S. (2015). Quantitative analysis for management (12th ed.). Pearson Education.

Suggested Unit Resources In order to access the following resources, click the links below. Read pp. 3–16 of the following study for an example of sampling, data collection, and data analysis in a mixed methods (qualitative and quantitative) study. Algan, E. K., & Ummanel, A. (2019). Toward sustainable schools: A mixed methods approach to investigating

distributed leadership, organizational happiness, and quality of work life in preschools. Sustainability, 11(19), 5489.

Read the following article for an additional example of data sampling, collection, and analysis in a quantitative study. Biddle, L., Menold, N., Benter, M., Nöst, S., Jahn, R., Ziegler, S., & Bozorgmehr, K. (2019, July). Health

monitoring among asylum seekers and refugees: A state-wide, cross-sectional, population-based study in Germany. Emerging Themes in Epidemiology, 16(1), 1–21. https://link.gale.com/apps/doc/A592920987/AONE?u=oran95108&sid=AONE&xid=aaed38c8

Read pp. 195–201 of the following study for an additional example of sampling, data collection, and data analysis in a qualitative study. Stoetzer, U., Bergman, P., Åborg, C., Johansson, G., Ahlberg, G., Parmsund, M., & Svartengren, M. (2014).

Organizational factors related to low levels of sickness absence in a representative set of Swedish companies. Work, 47(2), 193–205. https://libraryresources.columbiasouthern.edu/login?url=http://search.ebscohost.com/login.aspx?direc t=true&db=bsu&AN=94762593&site=ehost-live&scope=site