97%(104)101 out of 104 people found this document helpful
This preview shows page 20 - 24 out of 29 pages.
Lost-time-hours can be predicted by the following linear formula: Y = a + bXLost-time-hours = 273.4 + (-0.143)
21Sun Coast Research PaperEvaluationEvery new contract awarded to Sun Coast shall be trained in health and safety program. Data were collected from 223 contracts on training costs and lost-time hours. It would be useful to know how to predict lost hours of training, if training had been successful in reducing lost-time hours and, if so, how to predict lost-time hours from training expendituresMultiple RegressionsThe data were visually and statistically observed to determine if they fulfilled the necessary assumptions required for parameter testing. In addition, the graphic statistics below this confirm that skewness and kurtosis are within the acceptable ranges of -2 and +2. Mean, median, and mode as well as the normal distribution of data, closely aligned. These graphic statistics confirm that the information is suitable for data analysis using parameter statistical procedures.Regression StatisticsMultiple R0.601841822R Square0.362213579Adjusted R Square0.360083364Standard Error5.51856585Observations1503ANOVAdfSSMSFSignificanceFRegression525891.887845178.377569170.03614672.1289E-143Residual149745590.4898630.45456904Total150271482.3777CoefficientsStandardErrort StatP-valueLower 95%Upper 95%Intercept126.82245550.623820253203.29967630125.5988009128.0461101Frequency (Hz)-0.00111694.7551E-05-23.488460424.0652E-104-0.001210174-0.001023627Angle in Degrees0.0473423530.0373080691.2689574620.204653501-0.0258392880.120523993Chord Length-5.4953183352.927962181-1.8768406130.060734309-11.238662340.248025671
22Sun Coast Research PaperVelocity (Meters per Second)0.0832396340.0093001888.9503174361.02398E-180.0649968510.101482417Displacement-240.505908616.51902666-14.559326865.20583E-45-272.9088041-208.103013HypothesisH03 There is no statistically significant relationship between frequency, angle in degrees, chord length, velocity, and displacement and decibel level. HA3There is a statistically significant relationship between frequency, angle in degrees, chord length, velocity, and displacement and decibel level.Frequency, velocity, and displacement coefficient p-values < .05 (alpha), therefore, the null hypothesis (H03) is rejected and the alternative hypothesis (HA3) is accepted.Decibel level can be predicted by the following linear formula:Y = a + b1X1 + b2X2 +…+ bnXn dB = 126.8 + (-0.0011) (frequency) + (.083) (velocity) + (-240.5) (displacement)Independent Samples tTest: Hypothesis TestingH04:(null hypothesis) There is no statistically significant difference in mean scores between previous training and revised training.HA4:(alternative hypothesis) There is a statistically significant difference in mean scores between previous training and revised training.t-Test: Two-Sample Assuming Unequal VariancesGroup AGroup BMean69.7903225884.77419355Variance122.00449526.96456901Observations6262Hypothesized Mean Difference0df87t Stat-9.666557191P(T<=t) one-tail9.69914E-16t Critical one-tail1.662557349
23Sun Coast Research PaperP(T<=t) two-tail1.93983E-15t Critical two-tail1.987608282 Using t-test for independent sample p-value < .05 or 5% (alpha), we can see on the table below it shows that p-value= 1.93983E-15 < .05 or 5%. Therefore, the null hypothesis (H04) is rejected and the alternative hypothesis (HA4) is accepted because the null hypothesis is failed it doesn’t mean that the alternative hypothesis is not true. It only means that there is not enough evidence to reject the null hypothesis. However, the mean score between prior and revised training differs statistically significant.