Data-Driven-Decision-Making Practice Questions

A Pareto chart is a quality management tool that combines a bar chart and a line graph. The bars display the frequency or impact of problems in descending order, while the line graph shows the cumulative percentage. This format is specifically designed to highlight the "vital few" causes that contribute to the majority of problems, allowing teams to prioritize their improvement efforts on the most significant factors. This principle is often referred to as the 80/20 rule.

A. A run chart tracks a process performance metric over time to identify trends or patterns, not to categorize and prioritize problem causes.

B. A flowchart is a diagram that visually represents the sequence of steps and decisions in a process, not the significance of contributing factors.

D. A cause chart, often a cause-and-effect diagram, brainstorms potential causes for a problem but does not use data to sort them by significance.

1. Montgomery, D. C. (2013). Introduction to statistical quality control (7th ed.). John Wiley & Sons. Chapter 5, Section 5-2.4, "The Pareto Chart," pp. 193-194.

2. MIT OpenCourseWare. (2006). 16.660J Introduction to Lean Six Sigma Methods. IAP 2006. Lecture 2: Basic Quality Tools, slide 16, which describes the Pareto chart as a tool for prioritizing problems.

A simple index number is a statistical measure used to show the change in a variable, such as price or quantity, over time relative to a specific base period. The formula calculates this change by dividing the value in the current period by the value in the base period and then multiplying the result by 100. This converts the ratio into a percentage, making it easier to interpret, with the base period always having an index value of 100.

B. This formula is the inverse of the correct calculation and would incorrectly show a decrease when there is an increase, and vice versa.

C. This is an incorrect ratio (base/current) and omits the standard practice of multiplying by 100 to express the value as an index.

D. This provides the correct ratio but is not a complete index number, as it has not been multiplied by 100 for standardization.

1. Taylor, L. (n.d.). Lecture 10: Index Numbers. University of Cape Town, Department of Statistical Sciences. Retrieved from https://www.stats.uct.ac.za/stats/courses/sta1000/notes/Lec102013.pdf (See page 2, "Simple Price Index").

2. Illowsky, B., & Dean, S. (2018). Introductory Statistics. OpenStax, Rice University. Chapter 13, Appendix: Mathematical Phrases, Symbols, and Formulas. The general form of an index number is given as (current value / base period value) 100.

The classification of analytics into descriptive, predictive, and prescriptive categories is determined by the fundamental question it seeks to answer (its purpose) and the statistical or computational techniques it employs (its methods). Descriptive analytics summarizes past data to show what happened. Predictive analytics uses statistical models and forecasting techniques to understand what could happen. Prescriptive analytics uses optimization and simulation to advise on what should be done. The combination of purpose and method defines each category.

A. Sample size is a consideration for statistical validity, not a primary classifier of the type of analytics. Technique is only part of the answer.

B. Data validity and reliability are prerequisites for any meaningful analysis, not criteria for classifying the type of analysis being performed.

D. The software used is a tool for implementation; various software packages can perform all three types of analytics.

1. Davenport, T. H. (2013). Analytics 3.0. Harvard Business Review, 91(12), 64-72. (This article outlines the evolution and classification of analytics, emphasizing their purpose and capabilities).

2. Lustig, I., Dietrich, B., Johnson, C., & Dziekan, C. (2010). The Analytics Journey. Analytics Magazine, INFORMS. (This publication details the progression from descriptive to prescriptive analytics, defining each by its business purpose and enabling methods).

3. MIT Sloan School of Management. (n.d.). Business Analytics: From Data to Decisions. MIT OpenCourseWare. (Course materials often distinguish the three types of analytics based on the questions they answer and the techniques used, such as optimization for prescriptive analytics).

The firm's objectives span multiple business dimensions: internal operations (productivity), employee development (staff morale), customer relations (client satisfaction), and financial results. The Balanced Scorecard is a strategic performance management framework specifically designed to address this kind of multi-faceted goal setting. It provides a comprehensive view by measuring performance across four key perspectives: Financial, Customer, Internal Business Processes, and Learning & Growth. This structure directly aligns with and accommodates all the objectives stated by the professional services firm, ensuring a holistic approach to improvement.

A. The Net Promoter Score is a metric that only addresses the customer perspective (client satisfaction) and ignores the other stated goals.

B. Results-based management is a broad management strategy, not a specific performance tool for balancing diverse objectives like the Balanced Scorecard.

D. A KPI dashboard is a tool for visualizing metrics but lacks the underlying strategic framework of the Balanced Scorecard for selecting and balancing them.

1. Kaplan, R. S., & Norton, D. P. (1992). The Balanced Scorecard—Measures That Drive Performance. Harvard Business Review, 70(1), 71-79. (The original article outlining the four perspectives).

2. Stanford University. (n.d.). Balanced Scorecard. Stanford Graduate School of Business, Executive Education materials. Describes the framework as a tool for aligning business activities to the vision and strategy of the organization.

3. Nørreklit, H. (2000). The balance on the balanced scorecard—a critical analysis of some of its assumptions. Management Accounting Research, 11(1), 65-88. https://doi.org/10.1006/mare.1999.0121 (Provides an academic critique and analysis of the framework's structure).

The mean is sensitive to extreme values (outliers), while the median is resistant to them. When a dataset contains outliers, such as a few very high or very low salaries, the mean will be pulled in the direction of those outliers. The median, being the middle value, will be less affected. Therefore, a large difference between the mean and median salary is a strong indicator of a skewed distribution caused by outliers. This signals to the HR department that they need to assess these extreme values to understand the true salary landscape.

A. A large difference could be due to a few very high salaries, not necessarily a need for raises for the majority of employees.

B. A difference between the mean and median is a natural property of skewed data and does not, by itself, indicate a calculation error.

D. While outliers affect the standard deviation, the direct comparison of mean and median is the classic method for identifying skewness caused by these outliers.

1. University of California, Berkeley. (n.d.). Data 8: The Foundations of Data Science. Chapter 14.1: "Properties of the Mean." Retrieved from https://inferentialthinking.com/chapters/14/1/PropertiesoftheMean.html (This chapter explains how the mean is sensitive to outliers, unlike the median, and how their difference indicates skew).

2. Tufts University OpenCourseWare. (n.d.). BIO 132: Biostatistics. "Summarizing Data." (This course material describes how comparing the mean and median is a way to assess the symmetry of a distribution and detect the influence of outliers).

A scatter diagram (or scatter plot) is a fundamental tool in data analysis used to visualize and investigate the potential relationship between two continuous variables. By plotting the data points on a two-dimensional graph, one can observe patterns that suggest a positive correlation (as one variable increases, the other tends to increase), a negative correlation (as one increases, the other tends to decrease), or no correlation. It is an exploratory tool for identifying associations.

A. A prioritization matrix is a decision-making tool; a scatter diagram does not show its precedence or structure.

C. A bar chart or histogram is used to compare frequencies between groups, not a scatter diagram.

D. A t-test or ANOVA is used to compare means, often visualized with box plots or bar charts with error bars.

1. Montgomery, D. C. (2013). Introduction to Statistical Quality Control (7th ed.). Wiley. Chapter 5, Section 5.2.3 "The Scatter Diagram," states, "The scatter diagram is a useful tool for identifying a potential relationship between two variables." (p. 198).

2. MIT OpenCourseWare. (2009). 15.075J Statistical Thinking and Data Analysis. Fall 2009. Lecture 2: "Exploring and Describing Data." The lecture notes describe scatterplots as a primary method for visualizing the relationship between two quantitative variables.

3. Juran, J. M., & De Feo, J. A. (2010). Juran's Quality Handbook: The Complete Guide to Performance Excellence (6th ed.). McGraw-Hill. Chapter 15, "Basic Statistical Methods," describes the scatter diagram as a tool to "display what happens to one variable when another variable changes to test a theory that the two variables are related."

The primary objective of Six Sigma is to improve process quality by identifying and removing the causes of defects and minimizing variability. It aims for a process to be 99.99966% defect-free, which translates to no more than 3.4 defects per million opportunities (DPMO). This rigorous, data-driven goal is often described as approaching perfection in manufacturing and business processes. The methodology uses statistical methods to systematically improve processes toward this target level of quality and performance.

A. A balanced scorecard is a strategic performance management tool, not a primary objective of Six Sigma.

C. ISO 9000 is a separate set of quality management standards; Six Sigma's goal is not to establish them.

D. Six Sigma provides a methodology (e.g., DMAIC), but its ultimate objective is defect reduction, not just providing a roadmap.

1. Antony, J., & Banuelas, R. (2002). Key ingredients for the effective implementation of Six Sigma program. Measuring Business Excellence, 6(4), 20-27. https://doi.org/10.1108/13683040210451679. The article defines Six Sigma's goal as achieving a defect rate of 3.4 DPMO, which is a state of near-perfection (p. 21).

2. Schroeder, R. G., Linderman, K., Liedtke, C., & Choo, A. S. (2008). Six Sigma: Definition and underlying theory. Journal of Operations Management, 26(4), 536-554. https://doi.org/10.1016/j.jom.2007.06.007. The paper defines Six Sigma as a high-performance, data-driven approach to analyzing and solving root causes of business problems, with a goal of approaching zero defects (p. 536).

An appropriate management use of statistics is to gather and analyze data to gain insights and make informed decisions. Understanding customer demographics through descriptive statistics is a fundamental business application. This allows management to better understand their target market, tailor products and services, and develop effective marketing strategies. This process uses data to inform strategy, which is a core principle of data-driven decision-making.

B. Ordering product for a region based on one store is poor sampling. A single store is unlikely to be representative of an entire region, leading to inaccurate forecasts and inventory issues.

C. Using statistics to justify decisions after the fact is a misuse of data, often involving confirmation bias, rather than using data to inform the decision-making process itself.

D. The response rate of a survey indicates engagement but not substance. Decisions should be based on the analysis of the actual responses, not the percentage of people who responded.

1. Levine, D. M., & Szabat, K. A. (2021). Business Statistics: A First Course (8th ed.). Pearson. In Chapter 1, Section 1.1, the text outlines the use of statistics for summarizing business data, such as survey results on customer demographics, to draw conclusions and make better decisions.

2. Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., & Cochran, J. J. (2018). Modern Business Statistics with Microsoft Excel (6th ed.). Cengage Learning. Chapter 1, Section 1.4, discusses the role of descriptive statistics in summarizing data from customers to understand characteristics like location, gender, and age.

To design a cost-effective program, a government entity can apply benchmarking, a key data analytic approach. By analyzing data on the strategies, costs, and outcomes of recycling programs in other similar counties, the government can identify best practices and proven, efficient models. This comparative analysis allows them to adopt strategies that are most likely to succeed and be cost-effective, avoiding the trial-and-error costs of developing a program from scratch.

A. Analyzing budgetary impacts of contracts is a management task for an existing program, not a method for designing a new one to be cost-effective.

B. Aligning funds with employee numbers is an internal allocation tactic that does not guarantee the overall program's cost-effectiveness.

D. Researching audit performance relates to financial compliance and accountability, not the operational cost-effectiveness of a specific service like recycling.

1. Hatry, H. P. (2006). Performance Measurement: Getting Results (2nd ed.). The Urban Institute Press. Chapter 7, "Using Performance Information," explains how governments use comparative analysis (benchmarking) against other jurisdictions to identify best practices and improve program effectiveness and efficiency (pp. 151-155).

2. Kouzmin, A., Löffler, E., Klages, H., & Korac-Kakabadse, N. (1999). Benchmarking and performance measurement in public sectors. The International Journal of Public Sector Management, 12(2), 121-144. https://doi.org/10.1108/09513559910263462. This article discusses benchmarking as a tool for public sector organizations to learn from others to improve service delivery and cost-effectiveness.

In the design of experiments, "treatments" are the different procedures, conditions, or levels of a factor that are applied to the experimental units (the subjects of the study). The goal is to measure and compare the effects of these different treatments on a response variable. For example, in a medical study, the treatments might be a new drug, a standard drug, and a placebo.

A. Responses are the outcomes or measurements recorded after a treatment has been applied to an experimental unit. They are the effect, not the cause.

B. Inputs is a general term. In experimental design, the more precise terms are "factors" or "independent variables," which are manipulated via the treatments.

D. Experimental units are the subjects (e.g., people, animals, plots of land) to which the treatments are applied.

1. Montgomery, D. C. (2017). Design and Analysis of Experiments (9th ed.). John Wiley & Sons. Chapter 1, Section 1.2, "Basic Principles," defines treatments as the set of circumstances created for the experiment in order to observe their effect on the response.

2. The Pennsylvania State University. (n.d.). STAT 502: Analysis of Variance and Design of Experiments. Eberly College of Science. In Lesson 1.2, "Terminology," a treatment is defined as "a specific combination of factor levels whose effect is to be compared with other treatments."