Laserfiche WebLink
<br />The standard deviation was calculated for each of the abovementioned groups and was <br />used to describe the distribution of the data and to form a test statistic (such as at-test). <br />A normality test was also performed for each of these groups to test whether or not the <br />data was normally distributed. The results of the normality tests help dictate which <br />statistical methods would be most appropriate for the analysis. <br />Appendix E: Standard error results tables <br />Appendix F: Normality test description and results tables <br /> <br />Statistical Method: Although most of the data was not found to be normally distributed, <br />the statisticians came to the conclusion that since the sample sizes were greater than 30, it <br />would still be valid to use a one-sample t-test and a two-sample t-test. These tests are <br />fairly robust even in cases in which there is a violation of the normality assumption. <br /> <br />To determine the impacts of Slow the Flow Colorado, it was necessary to address <br />questions related to pre-inspection and post inspection data, including (i) what the <br />watering habits of each participant were in relation to ET rates during the pre and post <br />inspection years (ii) if the water usage was reduced from pre-inspection to post- <br />inspection as the result of the program. To answer these questions, the t-test was utilized <br />for the analysis of the data. <br /> <br />The first question (i) compares the means of percent +/- ET with a Imown value, which is <br />set to equal a household watering exactly at ET (0%). For this purpose, a two-tailed one <br />sample t-test was applied. For the second question (ii), a two-tailed, two sample paired t- <br />test was applied because the two samples (pre and post inspection) that were tested were <br />not independent by the structure of the experiment; the pre-inspection and post-inspection <br />observations were from the same subjects. <br />Appendix G: T-test description and results tables <br /> <br />Using the appropriate t-test the statisticians: <br />(1) Compared the Mean of the pre-inspection year with 0; <br />(2) Compared the Mean of the post-inspection year with 0; <br />(3) Compared the Means of the pre-inspection and post-inspection years; <br />(4) Compared the distribution of percent +/-ET for pre and post-inspection years; <br />(5) Found the difference in percent +/-ET for pre and post-inspection years; <br />(6) Found the percent difference for properties that watered above ET in the pre-inspection <br />year compared to the post-inspection year. <br />(7) Found the percent difference for properties that watered below ET in the pre-inspection <br />year compared to the post-inspection year. <br /> <br />. Zero represents a household watering exactly at ET. <br />. 2004 group (referred to as 2004AG): pre-inspection year was 2003 and post- <br />inspection year was 2005. Study group size== 208 households <br />. 2005 group (referred to as 2005AG): pre-inspection year was 2004 and post- <br />inspection year was 2006. Study group size== 645 households <br /> <br />25 <br />