Next simple regression is the preferred statistical procedure used to test

# Next simple regression is the preferred statistical

• 31
• 100% (26) 26 out of 26 people found this document helpful

This preview shows page 12 - 18 out of 31 pages.

Next, simple regression is the preferred statistical procedure used to test whether a relationship exists between lost-time hours and safety training expenditures. Additionally, multiple regression is the preferred statistical procedure to test the difference between frequency, angle in degrees, chord length, velocity, and displacement and dB levels. Moreover, the independent t - test is the preferred statistical procedure used to test where effectiveness lay between Group A and Group B training scores. Furthermore, the paired t -test is the preferred statistical procedure used to test the difference of blood lead levels at pre-exposure and post-exposure of remediation. Lastly, one-way analysis of variance (ANOVA) is the preferred statistical procedure used to test RESEARCH PROJECT 13 the difference of return on investment for air monitoring, soil remediation, water reclamation, and health and safety training projects. Data Analysis: Descriptive Statistics and Assumption Testing In this part of the study, I will describe the data that has been collected. The data are described both visually and statistically. Visually, the data is displayed in the form of graphs, histograms, and tables. This reveals the distribution of data, anomalies, and outliers. As well, descriptive statistics are used to test data to ensure the assumption of normality is met. Correlation: Descriptive Statistics and Assumption Testing Frequency distribution table. Histogram. Descriptive statistics table. RESEARCH PROJECT 14 Measurement scale. Ratio measurement scale, because it has absolute zero. Measure of central tendency. Mean is the measure of central tendency. Evaluation. Based on the descriptive statistics above, the mean, median, and mode values are similar, suggesting that data is consistent with a normal distribution. The kurtosis and skewness scores are between -2 and +2, which suggests satisfactory results to accept normal distribution. The assumptions for parametric statistical testing were met. Simple Regression: Descriptive Statistics and Assumption Testing Frequency distribution table. Histogram. RESEARCH PROJECT 15 Descriptive statistics table. Measurement scale. Ratio measurement scale, because it has absolute zero. Measure of central tendency. Mean is the measure of central tendency. Evaluation. Based on the descriptive statistics above, the mean, median, and mode values are similar, suggesting that data is approximately a normal distribution. The kurtosis and skewness scores are between -2 and +2, which suggests satisfactory results to accept normal distribution. The assumptions for parametric statistical testing were met. Multiple Regression: Descriptive Statistics and Assumption Testing Frequency distribution table. RESEARCH PROJECT 16 Histogram. Descriptive statistics table. RESEARCH PROJECT 17 Measurement scale. Interval measurement scale, because it has no unique origin (true zero).  • • • 