signature question

Research Work

Jeremy Harlow

University of Phoenix

I read this paper in Week 3. My comment was the lack of showing relevance to the project topic.

This week, the paper is being graded.

1

Research Work

Numerical Measures

Researchers use different types of numerical measures for data summarization. The percentage of data values in each data category is the most commonly used numerical measure for qualitative data analysis. The mean, median, mode, percentiles, range, and variance, as well as standard deviation are the primary measures for quantitative data analysis (Peck, Olsen & Devore, 2015). The mean, variance, and standard deviation will be used for data analysis. Covariance and coefficient correlation measures will also be used for data analysis. The mean, one of the major measures of central tendency, is the sum of data values divided by number of data values (Peck, Olsen & Devore, 2015). It will help in determining the data’s central location. The mean will be used because it includes all values in the data set during computation. In addition, the sum of deviations of each data value from the mean is normally zero. However, it is susceptible to outlier’s effect.

In addition to considering measures of location, it would be important to consider measures of variability. Measures of variability, also known as measures of dispersion, describe the spread in a given sample or population. A measure of spread will help in understanding how the mean represents the data. A large spread would mean that the mean does not represent the data because a large spread signifies large variations between individual scores (Peck, Olsen & Devore, 2015). Variance and standard deviation are two major measures of variability that will be used during this study. Variance is a useful measure of variability. According Holcomb (2016), a large variable will occur where the scores of group data are spread out. Conversely, the variance will be small if the scores of data are spread around the mean. The standard deviation, on the contrary, measures the spread of scores within a given set of data. It will be important to measure the population standard deviation since the study population all important values. To do so, it will be important to have the entire population or have a population sample. By using the standard deviation together with the mean, it will be possible to summarize research data.

Covariance and coefficient correlation measures will also help in determining relationship between dependent and independent variables. Bartolucci, Singh, and Bae (2015) indicate that covariance is a measure of linear relationship between two variables. Normally, a positive value indicates positive relationship while a negative value indicates a negative relationship. Correlation coefficient measures the strengths as well as the direction of relationship existing between two variables measured on interval scale. This numerical measure will help in understanding whether there is an association between worker job satisfaction and worker performance. Correlation co-efficient, especially Pearson’s correlation seeks to identify a line of best fit through two data variables, with the Pearson correlation, r, showing how well data points fit the line of best fit.

Graphical Measures

There are different types of graphs, charts, pictorial, and table methods that researchers can use to enhance their understanding of research variables as well as the relationship existing between such variables. Graphical, tabular, and pictorial methods of data presentation provide visual aspect of data. Some of the commonly used measures include: histograms, scatter plots, charts, and tables (Bartolucci, Singh & Bae, 2015). Graphs, pie charts, tables, and scatter graphs will be used in this study for data representation and analysis. Graphical displays will help in completing tabular presentation of data variables. Holcomb (2016) indicates that graphs helps in determining data patterns, while tables help in analyzing large amounts of data characterized with high level of numerical detail. While there are various types of graphs such as bar graphs, histograms, and a box and whisker plot, cumulative frequency plot will be used to display distribution of data variables or values. Duquia et al. (2014) indicate that cumulative frequency graphs help in observing and determining the number of data values lying below or above a given range of data set. In addition, it helps in understanding how data values in a certain data set vary.

Frequency distribution of data variables will also be displayed in tables. A table helps in providing a broad range of information on gathered data. When evaluating frequency distribution of continuous data variables using tables, it is important to transforms data variables into groups with same size. A contingency table can help with investigation of association existing between categorical data variables (Duquia et al., 2014). It will help in analyzing the relationship between independent and dependent variables.

The same information contained in the tables will be represented in form of pie chart. A pie chart indicates the association of components of the whole. A scatter plot will also help in showing the relationship between employee job satisfaction and employee performance. According to Holcomb (2016), a scatter plot can help in determining whether two variables have a linear relationships and detecting any outliers. It also graphically represents an association between two continuous data variables.

References

Bartolucci, A., Singh, K. P., & Bae, S. (2015). Introduction to statistical analysis of laboratory data (1st ed.). Hoboken, NJ: John Wiley & Sons.

Duquia, R. P., Bastos, J. L., Bonamigo, R. R., González-Chica, D. A., & Martínez-Mesa, J. (2014). Presenting data in tables and charts. Anais brasileiros de dermatologia, 89(2), 280-285.

Holcomb, Z. C. (2016). Fundamentals of descriptive statistics. Abington Thames, UK: Routledge.

Peck, R., Olsen, C., & Devore, J. L. (2015). Introduction to statistics and data analysis (5th ed.). Boston, MA: Cengage Learning.