Effect of Salinity on Cell Diameter - Report Example

Effect of salinity on cell diameter. This section gives the result of the investigation of the effect of salinity on cell diameter when two cultures are used. One of the cultures was grown at a salinity of 1 while the other was grown at a salinity of 30. The diameter of 25 cells in the culture were measured. The hypothesis of the test were HO There is no difference in the cell diameters at the two levels of salinity H1 There is a difference in the cell diameters at the two levels of salinity Test of normality and Homogeneity of Variance To find out whether a parametric test could be performed using the data normality test and homogeneity of variance were performed with the results being as in table 1 and table 2. From table 5 Shapiro –Wilk section, it can be seen that the data had a p value of more than 0.05 in both salinity 1 and salinity 30. From table 2 it can also be seen that based on mean the homogeneity test indicate that a t-test can be used as the p-value is more that 0.05. The mean cell diameter for salinity 1 was 13.5 and salinity 30 was 12.4 as can be seen in figure 1. Figure 1. Mean diameters for the two salinity levels. TABLE 1 Tests of Normality level of salinity Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. two level salinity sal1 .101 25 .200* .963 25 .475 sal30 .129 25 .200* .960 25 .405 a. Lilliefors Significance Correction *. This is a lower bound of the true significance. Table 2 Test of Homogeneity of Variance Levene Statistic df1 df2 Sig. two level salinity Based on Mean 2.656 1 48 .110 Based on Median 2.632 1 48 .111 Based on Median and with adjusted df 2.632 1 41.672 .112 Based on trimmed mean 2.659 1 48 .110 T-test A t-test was performed in order to find out more about the difference in diameters at the two salinity levels with the results being as seen in table 3. From the table it can be seen that there was a significant difference in the diameters at the two levels of salinity with a p-value at equal variances was 0.005 with a t value of 3.103. The table also shows that there was a mean difference of 1.0846. From this results we reject the null hypothesis while we accept the alternative hypothesis. Table 3 t-test at two salinity levels Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference Lower Upper salinity Equal variances assumed 4.855 .037 3.103 24 .005 1.0846 .3496 .3631 1.8061 Equal variances not assumed 3.103 17.508 .006 1.0846 .3496 .3487 1.8205 Identifying glass type using This test is used in identifying the glass type to which the two sample of glass belong to. From table 4 the mean values of the different compositions of glass in each glass type has been given. From the table it can be observed that milk bottle has the highest value of Na2O with a value of 16.33 while the least was 5.8325which was recorded in pyrex utensil. The glass type with the highest average value for SiO2 was in pyrex utensil glass type with a value of 90.16 while the least average value of 74.205 was recorded in milk bottle glass type. The other components in each glass type are as illustrated in the table. The canonical function has been used in locating the centroid each group of data. From the canonical descriminant function it can be seen that one of the unknown is much closer to pyrex utensil the other is close to wine bottle. The classification results (Table 5) give the predicted results where there is a confirmation that one of ungrouped cases is a pyrex utensil while the other is wine bottle. To identify the particular case it is important to revisit the data view after running the test. It is clear from the data view that the first extra column which has been generated gives the type of glass type of each ungrouped data. From the column it is found that the first ungrouped case which had a Na2O value of 2.55 was the one that was classified as being a pyrex utensil as it had assigned 2 in the column. The second ungrouped case which had a Na2O value of 10.22 was classified as wine bottle as it was assigned 4 in the column. Figure 2 Table 5 Classification Resultsa Glass_Type Predicted Group Membership Total Milk Bottle Pyrex Utensil Window Pane Wine Bottle Original Count Milk Bottle 4 0 0 0 4 Pyrex Utensil 0 4 0 0 4 Window Pane 0 0 4 0 4 Wine Bottle 0 0 0 4 4 Ungrouped cases 0 1 0 1 2 % Milk Bottle 100.0 .0 .0 .0 100.0 Pyrex Utensil .0 100.0 .0 .0 100.0 Window Pane .0 .0 100.0 .0 100.0 Wine Bottle .0 .0 .0 100.0 100.0 Ungrouped cases .0 50.0 .0 50.0 100.0 a. 100.0% of original grouped cases correctly classified. Forensic Anthropology – Stature and anthropometry This exercise involves investigating the relationship between various variables that were measured. To investigate the relationships a correlation test was done between the various variables. Test of normality and Homogeneity of Variance To find out whether a parametric test could be performed using the data normality test and homogeneity of variance were performed with the results being as in table 6 and table 7. From table 6 Shapiro –Wilk section it is that the data had a p value of more than 0.05 in both male and female for height (standing) and weight. There was also a p-value of more than 0.05 in male height (sitting). In the other variables the p-value was less than 0.05. In the test for homogeneity based on mean p-values of more that 0.05 were recorded in Height (standing), weight and height (elbow) but we had a p-value less than 0.05 in height (sitting) Table 6: Tests of Normality Male/Female Kolmogorov-Smirnova Shapiro-Wilk Statistic df Sig. Statistic df Sig. Height (Standing) cms Female .098 25 .200* .964 25 .495 Male .099 13 .200* .984 13 .993 Weight (kgs) Female .142 25 .200* .930 25 .087 Male .210 13 .120 .932 13 .366 Height (Elbow) cms Female .296 25 .000 .636 25 .000 Male .213 13 .111 .857 13 .035 Height (Sitting) cms Female .341 25 .000 .685 25 .000 Male .188 13 .200* .925 13 .295 a. Lilliefors Significance Correction *. This is a lower bound of the true significance. Table 7: Test of Homogeneity of Variance Levene Statistic df1 df2 Sig. Height (Standing) cms Based on Mean .770 1 36 .386 Based on Median .663 1 36 .421 Based on Median and with adjusted df .663 1 33.686 .421 Based on trimmed mean .755 1 36 .391 Weight (kgs) Based on Mean 2.809 1 36 .102 Based on Median 2.682 1 36 .110 Based on Median and with adjusted df 2.682 1 28.864 .112 Based on trimmed mean 2.768 1 36 .105 Height (Elbow) cms Based on Mean .262 1 36 .612 Based on Median .257 1 36 .616 Based on Median and with adjusted df .257 1 29.756 .616 Based on trimmed mean .296 1 36 .590 Height (Sitting) cms Based on Mean 23.079 1 36 .000 Based on Median 4.434 1 36 .042 Based on Median and with adjusted df 4.434 1 24.317 .046 Based on trimmed mean 18.278 1 36 .000 Relationship between weight and height One of the investigation which was done on the variables was to establish if there was a correlation between the weight and height of the participants. The hypothesis for this test was HO There is no correlation between height (standing) and weight of the participants H1 There is a positive correlation between height (standing) and weight of the participants The result of the test was as shown in table 8. From the table it can be seen that a significant positive correlation between the two variables where the correlation value is 0.658 at p0.01. This is a clear indication that the null hypothesis is true and thus should be accepted while the alternative hypothesis is false and should be rejected. Relationship between elbow height and height (standing) The correlation test was performed between the elbow height of the participants and the height (standing). The hypothesis for the investigation was HO There is no correlation between elbow height and height (standing) of the participants H1 There is a positive correlation between elbow height and height (standing) of the participants From table 8 it can be observed that a significant positive correlation between the two variables where the correlation value is 0.626 at p Read More