91%(23)21 out of 23 people found this document helpful
This preview shows page 6 - 8 out of 12 pages.
Consider research on the level of education of the employees in a company. If the company is large, a sample of the company could be used. A quick survey could be sent to employees from various departments of the company asking to indicate the level of education completed as high school diploma, associate’s degree, bachelor’s degree, master’s degree, or Ph.D. The results of the study could be tallied and displayed on a bar chart or pie chart.Explaining the DataTo explain the bar chart, the report should indicate a sample of 40 employeesfrom various departments was surveyed. The results indicate that 7.5% completed high school only, 17.5% completed an associate’s degree, 70% completed a bachelor’s degree, 5% completed a master’s degree, and 0% completed a Ph.D. program. Although the results provide some insight for the entire company, it cannot be assumed that, for example, 5% of the employees in the company have a master’s degree.The measure of the center of a data set typically provides a clear picture of the average of a data set. However, if an outlier or a significantly higher or lower value is present in the data set, the average is skewed. Consider the average salary for a department in the organization. The majority of the employees in the department are young with 3–5 years of experience. The manager’s salary may skew the results causing the mean salary to be higher than it truly is.
Example: Salaries (in thousands): 41, 40, 39, 39, 43, 50, 44, 42, 43, 85.The mean salary for the sample is $46,600, and the median salary is $42,500. A difference in the mean and median of $4,100 is significant. Calculating the mean with the outlier of 85 reduces the mean to $42,333. Submitting a report with the findings of the mean salary for the department of $46,600 without the explanation of the outlier could have a negative impact of the decision involving the department. Although the research is accurate and represents the mean salary, presenting the results with and without the outlier is more accurate.When reporting a correlation from correlation research, the correlation coefficient should be presented along with the correlation type. A positive correlation can have a coefficient ranging from close to 0 to 1.0. Although the range seems insignificant, it is not.The scatter plot presents a positive correlation between experience and success in negotiating. To strengthen the findings, the correlation coefficient is 0.79, which is approaching 1.0. Thus, this confirms a strong positive correlation.Forinferential statistics, research involves a hypothesis formed by an inference for the population based on the sample. As with descriptive, a report of the research can be misunderstood if the research basis is not presented. The sample and population should be defined along with the descriptive statistic involved in the hypothesis. Then, the level of significanceshould be included to indicate the chance of a type I error. When stating the conclusion on the hypothesis test, the results should clearly state the null hypothesis.