A Type I error is a greater concern than a Type II error. Agree or disagree with this statement and use examples to support your position. Choose a classmate with an opposing view and try to respectfu

MBA 5652, Research Methods 1 Cou rse Learning Outcomes for Unit II Upon completion of this unit, students should be able to: 2. Examine the developmental components of the research process. 2.1 Differentiate between research questions and hypotheses. 2.2 Develop an introduction to a research project that includes an explanation of the research problems, the purpose of the study, research questions, and hypotheses about a business dilemma. Course/Unit Learning Outcomes Learning Activity 2.1 Unit Lesson Chapter 3 Chapter 7 Article: “ Hypothesis Testing, Type I and Type II Errors ” Unit II Scholarly Activity 2.2 Chapter 5 Unit II Scholarly Activity Reading Assignment Chapter 3: The Use of Theory, pp. 49 –52 Chapter 5: The Introduction, pp. 101 –110 Chapter 7: Research Questions and Hypotheses, pp. 136 –141 In order to access the following resource, click the link below: Banerjee, A., Chitnis, U. B., Jadhav , S. L., Bhawalkar, J. S., & Chaudhury, S. (2009). Hypothesis testing , type I and type II errors. Industrial Psychiatry Journal, 18 (2) , 127 –131. Retrieved from https://www.ncbi.nlm.nih.g ov/pmc/articles/PMC2996198/ Unit Lesson This unit covers the topic of research questions and hypotheses. Research questions and hypotheses are closely related since they both function to describe, and ultimately solve, the management problem(s). The following will explain these foundational concepts of business research. Research Questions To solve the business problem (s), it is necessary to answer the resea rch question(s). For example, consider a scenario where a U.S. company needs to make a large capital expenditure for a piece of manufacturing equipment. There are five models offered by five different vendors, and all might be suitable replacements for the obsolete equipment. The business problem is that they do not have enough information to determine which model is the best investment. Given the expensive price tag, the company wants to make the right decision to replace the obsolete equipment. The wrong decision would be quite costly. The company’s research consultant is commissioned to do a comparison of the five models of equipment:

Models A, B, C, D, and E. The consultant’s job is to collect and test data on the five models so that the client UNIT II STUDY GUIDE Research Questions and Hypotheses MBA 5652, Research Methods 2 UNIT x STUDY GUIDE Title can make an evidence -based decision to reduce risk. There are several variables to consider including production capacity, cost of operation, and error rate. The research question is this: Which model will provide the most productivity? W hen this question is answered, the business problem is solved. Before the question is answered, however, it must be restated as a hypothesis and statistically tested. Hypotheses A hypothesis is an explicit statement about relationships or differences between variables or groups that is speculative, yet unsubstantiated (Cooper & Schindler, 2011). The consultant’s hypothesis, which is related to both the problem and research question, would be a prediction about the differences in productivity between Models A, B, C, D, and E. This would be a causal -comparative study. In this scenario, model type is the independent variable (also known as predictor variable), while production capacity, cost of operation, and error rate are the dependent variables (also known as outcome variab les). In this causal -comparative scenario, the company happens to be hypothesizing about differences between models since finding differences would help answer the research question. As mentioned previously, hypotheses may also speculate about relationshi ps between variables. For example, assume the research question is this: W hich model has the highest unit output relative to cost per unit? Since cost per unit would decrease as production increases, the relationship between these variables, or correlation , would allow the company to compare models and make a sound business decision based on the strengths of the correlations.

Using a hypothesis about relationships, the company would ideally like to determine the model with the greatest negative correlation between output and unit costs. In other words, as production increases, unit costs decrease. Stating hypotheses in inferential statistics is done in a specific manner. It is customary (and sometimes referred to as traditional) to state both a null hypothe sis and an alternative hypothesis. A null hypothesis is a statement where no statistically significant relationship or statistically significant difference occurs between the variables or groups under study (Creswell & Creswell, 2018). Null hypothesis: Th ere are no statistically significant differences in production capacity, cost of operation, and error rate between Models A, B, C, D, and E. Alternative hypothesis: There are statistically significant differences in production capacity, cost of operation, and error rate between Models A, B, C, D, and E. There are several unique things to notice about these hypotheses. The first thing is that the alternative hypothesis is the opposite of the null hypothesis. The null hypothesis should be viewed as the stat us quo. The status quo in the scenario is that no differences exist between models. The purpose of conducting the research is to determine if statistical differences do indeed exist. This is done by statistically analyzing the data to determine if the stat us quo, or the null hypothesis, can be rejected (Swanson & Holton, 2005). It is important to note that when engaging in hypothesis testing, the researcher can never make the statement that the hypothesis is proven or disproven. Research results simply allo w the researcher to either accept or reject the null hypotheses. If research provided proof, there would be no reason to continue to conduct scientific inquiry. For example, the atom was once considered the smallest component of matter. Now, through scienc e and hypothesis testing, atomic theory has moved to string theory. If researchers accepted the hypothesis that the atom was the smallest unit of matter as proof, we would know nothing about strings, vibrations, and their relationship to matter. Similarly, hypotheses about strings are not proven but accepted or rejected. The researcher should always speak in terms of either rejecting or accepting the null and alternative hypotheses. More specifically, it should be explicitly stated that, based on the stati stical results, either the null hypothesis is rejected and the alternative hypothesis accepted, or the null hypothesis is accepted and the alternative hypothesis is rejected. It is an impossibility to either reject both the null and alternative hypotheses or accept both the null and alternative hypotheses since they are mutually exclusive events. Statistical Significance Notice in the null and alternative hypotheses stated above the use of the phrase statistically significant . Statistical procedures test for significant differences or significant relationships between variables. Before making decisions about Models A, B, C, D, and E, the researcher must have a certain level of confidence in MBA 5652, Research Methods 3 UNIT x STUDY GUIDE Title the results from the statistical procedure. The information needed to interpret statistical results includes the significance level (also known as alpha), a calculated statistical value, and the calculated statistic’s corresponding p value (probability value). The alpha is a benchmark that is set by the researcher. In bu siness and social science research, the alpha is customarily set at .05 (5%). An alpha of .05 is the normally accepted cut -off for rejecting the null hypothesis and accepting the alternative hypothesis (Field, 2005; Norusis, 2008). After the statistical procedure has been completed, the calculated statistic’s p value is then compared to the alpha. If the p value is less than .05, the null hypothesis is rejected, and the alternative hypothesis is accepted. If the p value is greater th an .05, the null hypothesis is accepted, and the alternative hypothesis is rejected. Having a p value <.05 is the same as saying that there is less than a 5% probability that the results were due purely to chance (Field, 2005). Conversely, the researcher c an be at least 95% confident that the test results represent the true effect being measured. Having 95% confidence still leaves a small chance of error. Even if the test results are statistically significant at a p value <.05, suggesting a greater than 95 % confidence that the calculated value represents the true effect, there is still a small possibility that our results could have been due to random chance. If these small possibilities occur, they are known as Type I and Type II errors. A Type I error is when the null hypothesis is rejected when, in fact, it should have been accepted; this is a false positive. A Type II error is when the null hypothesis is accepted when, in fact, it should have been rejected; this is a false negative. After testing the ab ove hypotheses for statistically significant differences in production capacity, cost of operation, and error rate between equipment Models A, B, C, D, and E, the results can be interpreted and recommendations made. For example, if the statistical results indicate that Model A has statistically higher production capacity, lower cost of operation, and lower error rates versus Models B, C, D, and E, an evidence -based decision could be made to purchase Model A. Hypothesis Testing Steps The basic steps for co mpleting a test of statistical significance are listed below (Cooper & Schindler, 2011). 1. State the null and alternative hypotheses. 2. Choose a statistical test (e.g., correlation, regression, ttest, ANOVA, etc.) 3. Select the desired alpha (significance level). Before running the statistical test, it must be decided how much risk is tolerated. In business research, a .05 significance level, or level of risk, is customary.

4. Calculate the statistic and associated pvalue. This is commonly done via statistical software . 5. Interpret the test. If the pvalue for the calculated statistic is less than .05, the null hypothesis is rejected, and the alternative hypothesis is accepted. If the pvalue is greater than .05, the null hypothesis is accepted, and the alternative hypothesis is rejected. MBA 5652, Research Methods 4 UNIT x STUDY GUIDE Title Characteristics of Good Hypotheses Good hypotheses are declarative, speculative, brief, and testable (Cooper & Schindler, 2011). Hypotheses examples: Both the nu ll and alternative hypotheses should be stated. Example one: Statistical test looking for a difference (causal -comparative) between the same x variable in different groups. Ho1 : There is no statistically significant difference in x between Group A and Group B. Ha1 : There is a statistically significant difference in x between Group A and Group B. Example two: Statistical test looking for a relationship (correlation) between x and y variables. Ho2: There is no statistically significant relationship between x and y. Ha2: There is a statistically significant relationship between x and y. As shown above, the null and alternative hypotheses should both be stated when formulating hypotheses.

Similarly, both the null and alternative hypotheses should be stated when interpreting test results. It should be clear which is accepted and which is rejected when evaluating the statistical value and its corresponding p value . References Cooper, D. R., & Schind ler, P. S. (2011). Business research methods (11th ed.). New York, NY: McGraw - Hill/Irwin. Creswell, J. W ., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). Thousand Oaks, CA: Sage. Field, A. (2 005). Discovering stats using SPSS (2nd ed.). London, England: Sage. Norusis, M. J. (2008). SPSS 16.0 guide to data analysis (2nd ed.) . Upper Saddle River, NJ: Prentice Hall. Swanson, R. A., & Holton, E. F., III (Eds.). (2005). Research in organizations: Foundations and methods of inquiry . San Francisco, CA: Berrett -Koehler.