Unit VI Research Paper

RCH 8301, Quantitative Research Methods 1

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

6. Create research questions appropriate for a selected research method and design. 6.1 Develop a research topic, and include appropriate research questions.

7. Formulate hypotheses appropriate for a selected research method and design.

7.1 Design hypotheses that are suitable for a selected research method and design.

Course/Unit Learning Outcomes

Learning Activity

6.1, 7.1

Unit Lesson Chapter 18, pp. 318–331 Chapter 19, pp. 334–346 Unit VI Research Paper

Required Unit Resources Chapter 18: General Design Classifications for Selection of Difference Statistical Methods, pp. 318–331 Chapter 19: Selection of Appropriate Statistical Methods: Integration of Design and Analysis, pp. 334–346

Unit Lesson

General Design Classifications

Researchers must think critically about the type of information that is needed to address a research problem, and then researchers must make sure that the overall research problem will be adequately addressed. If they do not do this, they may reach conclusions that are unconvincing, and the overall validity of the study may be questioned. In this unit, we will focus on general design classifications, which will help us determine the proper format and statistical approach to use.

Present-day statistics offer the basis for inference in various research studies. In the various differential methods for statistical analysis, there are procedures called general design classifications. These general design classifications are between-group design, within-subject design (repeated measures design), and mixed design. However, the focus in this unit is to distinguish the general design classification for comparative research, experimental, and quasi-experimental approaches with the aim of understanding the selection of appropriate statistical methods. The study design is considered a general plan that is used in setting up and testing a research question or a specific hypothesis (Thompson & Panacek, 2006). This implies that the research design directs the researcher on the who, when, what, and how regarding how the study project is conducted. Consequently, the general design classifications are important in the determination of

the appropriate statistical methods that the researcher adopts in the data analysis stage. Therefore, it is a necessity in the randomized experimental, comparative, and quasi-experimental approaches that all of the

UNIT VI STUDY GUIDE

Selection of Appropriate Statistical Measures

(Alexmillos, n.d.)

RCH 8301, Quantitative Research Methods 2

UNIT x STUDY GUIDE

Title

designs be appropriately fitted in the categories (i.e., between-group design, within-subject design, and mixed design). The between-group design refers to a design whereby each participant in a research project is in only one group or condition (Morgan et al., 2002). Accordingly, this design requires that each participant in the research study receive only one of the two conditions set in the experiment. For example, in a study where the effects of high temperature on the growth of a plant might have two groups of the independent variable (retarded growth or improved growth), each plant will only achieve one condition. Thus, the choice of study participants (sample size) will be influenced by these groups, where each group will have the number of participants doubled. In a within-subject design (repeated measures design), which is the opposite of the between-group design, a general design classification is realized. According to Morgan et al., a within-subject design is where each participant in the research project receives all of the conditions. This implies that each participant in the study experiences all levels of an independent variable to complete the study. For example, in a study where a drug is tested among children to establish the outcome between the two sets of doses (current and new medications as independent variables), the within-subject design requires that each participant in the study receive both medications; therefore, a number of symptoms would be measured on both of the independent variables. Furthermore, in this design, the number of participants is not affected by the variables used like it is in the case of between-group design since each participant receives all or both conditions of the independent variable in the study. Therefore, the within-subject design is referred to as a repeated measures design because of the experimental conditions where each participant is assessed more than once depending on the research conditions. Despite the existing advantages of the within-subject design, such as a reduction in the error variance and a reduction in the number of participants, this design is considered less appropriate compared to the between-group design. Its inappropriateness is derived from the possibility of participants having carryover effects, especially in studies where the change over time in the response to medication (example provided earlier) is an independent variable. Otherwise, both the between-group design and the within-subject design have a similarity in the number of independent variables considered, which is only one. A mixed design has more than one between-group independent variable as well as one within-subject independent variable. This implies that this design has at least two independent variables studied. Consider the aforementioned experiment where the effects of high temperature on the plant growth are to be investigated; in the mixed design, an additional independent variable (between-group) will be required, thus identifying this as a mixed design. In this case, the variety of the plant may be introduced as the additional independent variable to study the effects of high temperature on the growth of the plants. In the design considerations for a mixed design, there is the need for the researcher to appreciate the dimensions of the design (e.g., issues of validity). The design dimensions in the mixed research include the theoretical drive, purpose, timing, design complexity, and planned design (Schoonenboom & Johnson, 2017). Thus, in the mixed design, both qualitative and quantitative approaches are considered in the use of theory, the use of logic, the purpose of the results, the view of objectivity, the sampling of strategies, and the choice of statistical methods for data analysis.

Selection of Appropriate Statistical Methods

There are various aspects that must be considered when selecting an appropriate statistical method in the design and analysis of a research project. When selecting a statistical method, the concepts that must be considered include the research approaches and questions, dependent and independent variables, design classification, statistical assumptions, and the levels of measurement. The first step toward selecting a statistical method is defining the level of measurement for all of the variables (nominal, interval, ratio, or ordinal level) that are studied and included in the analysis. However, the use of tables is also effective when trying to select the appropriate statistics for the design and analysis of the result findings. For example, in the common single comparison tests, the dependent variable (scale) and the independent variable (nominal) would require different parametric tests and non-parametric tests, such as the independent-samples t-test and the Wilcoxon rank-sum test, respectively.

RCH 8301, Quantitative Research Methods 3

UNIT x STUDY GUIDE

Title

To arrive at the most suitable statistical method, it is recommended to identify whether the research question focuses on the association or the difference between variables and to identify the number of independent variables in the study. Moreover, the aspects of general design classifications (between-group, within-subject, and mixed designs) come into play. By using a schematic diagram, which describes the purpose, approach, type of question, and general type of statistics, it can be useful in helping one identify and select the appropriate statistical method that suits the research project. For example, one is required to distinguish between the relationship between variables (experimental or non- experimental) and thereby use the variables to identify the specific approach to be adopted in the study (randomized or quasi for an experimental approach and

comparative or associational for a non-experimental approach). Having identified the specific purpose for the variables, one would then identify the type of questions (e.g., difference for an experimental approach and associational or descriptive for a non-experimental approach), one would then determine the general type of statistic to be used. For example, difference inferential statistics would be used for the difference type of question, and associational inferential or descriptive statistics would be used for associational and descriptive statistics for descriptive-type questions. For example, difference inferential statistics use the t-test and analysis of variance (ANOVA); descriptive statistics use histograms, percentages, and means; and associational inferential statistics use correlation and regression (Gliner et al., 2017). As covered in the readings for this unit, selection of the appropriate statistical method requires good judgement. Since each research study is different, the most suitable research design and statistical analysis must be chosen.

References

Alexmillos. (n.d.). Business icons and target infographics (ID 64597036) [Illustration]. Dreamstime. https://www.dreamstime.com/stock-illustration-business-icons-target-infographics-illustration-design- graphic-image64597036

Gliner, J. A., Morgan, G. A., & Leech, N. L. (2017). Research methods in applied settings: An integrated

approach to design and analysis (3rd ed.). Routledge. Morgan, G. A., Gliner, J. A., & Harmon, R. J. (2002). General design classifications. Journal of the American

Academy of Child and Adolescent Psychiatry, 41(2), 226–228. https://www.jaacap.org/article/S0890- 8567(09)60667-5/fulltext

Schoonenboom, J., & Johnson, R. B. (2017). How to construct a mixed methods research design. Cologne

Journal of Sociology and Social Psychology, 69,107–131. https://link.springer.com/article/10.1007/s11577-017-0454-1

Tashatuvango (n.d.). Data analysis on white-golden compass (ID 44076262) [Illustration]. Dreamstime.

https://www.dreamstime.com/stock-illustration-data-analysis-white-golden-compass-needle-field- pointing-image44076262

Thompson, C. B., & Panacek, E. A. (2006). Research study designs: Experimental and quasi-experimental.

Air Medical Journal, 25(6), 242–246. https://www.airmedicaljournal.com/article/S1067- 991X(06)00286-0/abstract

(Tashatuvango, n.d.)

RCH 8301, Quantitative Research Methods 4

UNIT x STUDY GUIDE

Title

Learning Activities (Nongraded) Nongraded Learning Activities are provided to aid students in their course of study. You do not have to submit them. If you have questions, contact your instructor for further guidance and information. Review the “Interpretation Questions” and “Application Problems” at the end of Chapters 18 and 19.