MARGARET MCKOSKY
Appendix 6 Blog Documentation
5.7 TESTING OF HYPOTHESES
The hypotheses for this study were tested using suitable statistical tools and method. The results of the hypotheses were discussed below.
Hypothesis 1: There is no significant difference among the leachate, soil, plants
Factors of Interest: 5 factors are to be used for the hypothesis which include sample A, B, C, D and Control, based on that a confirmation of conformity of One-Way ANOVA (Test of basic assumptions) conducted.
Confirmation of conformity of One-Way ANOVA (Test of Basic Assumptions) Table 5.26: Normality Test
Factors P-value Remark
Sample A 0.005 Not normally distributed Sample B 0.003 Not normally distributed Sample C 0.005 Not normally distributed Sample D 0.005 Not normally distributed Control 0.003 Not normally distributed
Conclusion
The P-values are less than 0.05 which implies the observations for all the factors of interest are not normally distributed. This violated the first assumption of One-Way ANOVA and the appropriate test Statistic is Kruskal- Wallis Test.
Test of Hypothesis using Kruskal-Wallis test at 5%.
Kruskal-Wallis Test: Observations versus Factors
Kruskal-Wallis Test on Observations Factors N Median Ave Rank Z 1 15 6.500 43.2 1.03 2 15 3.900 40.3 0.46 3 15 2.500 36.7 -0.26 4 15 2.300 34.0 -0.79 5 15 2.760 35.8 -0.44 Overall 75 38.0
H = 1.74 DF = 4 P = 0.003
H = 1.74 DF = 4 P = 0.003 (adjusted for ties)
Conclusion
The P-value of the test is less than 0.05 which is an indication of existence of enough evidence to reject the null hypothesis and conclude that there is significant difference among the leachate, soil, plants and water of the study area.
(12...n) .
Further Test for Possible Classification of factors was done which is the Cluster Analysis.
Cluster Analysis of Variables: A, B, C, D, and CONTROL Correlation Coefficient Distance, Average Linkage
Amalgamation Steps
Number of obs.
Number of Similarity Distance Clusters New in new Step clusters level level joined cluster cluster 1 4 99.9947 0.000107 1 2 1 2 2 3 99.9904 0.000193 3 4 3 2 3 2 99.7702 0.004597 3 5 3 3 4 1 87.8972 0.242055 1 3 1 5 Final Partition
Cluster 1
A B Cluster 2
C D CONTROL
CONTROL D
C B
A 87.90
91.93
95.97
100.00
Variables
Similarity
Dendrogram
Average Linkage, Correlation Coefficient Distance
From the output of cluster analysis, two groups were formed; Sample A and Sample B while the second group is Sample C, Sample D and the Control.
This can be interested as no significant difference between samples A and B but samples C, D and Control are significantly different from A, B.
Testing of Hypothesis 2
Null hypothesis: The hydrodynamic solute dispersion trend of the heavy metals in soil and water in the study area do not significantly differ.
Alternative hypothesis: The hydrodynamic solute dispersion trend of the heavy metals in soil and water samples in the study area differ significantly.
Test Statistic: Student T-Test was used since we have two variable which is soil and water.
Table 5.27: Test of significant difference of Lead
t-Test: Two-Sample Assuming Unequal
Variances
In Soil In water
Mean 2.064833333 0.040333333
Variance 0.002974167 0.001728267
Observations 6 6
Hypothesized Mean
Difference 0
Df 9
t Stat 72.31561479
P(T<=t) one-tail 4.67409E-14 t Critical one-tail 1.833112923 P(T<=t) two-tail 9.34818E-14 t Critical two-tail 2.262157158
Discussion
In terms of Lead, the presence of the chemical in soil is significantly different from that of water with that of soil being higher than that of water, comparing the mean values. The P-value of the test is less than 0.05 which is an indication of the existence of enough evidence to reject the null hypothesis.
Using the calculated value and the table value, the calculated value is 72.32 and the table value is 2.62. Since the calculated value is greater than the table value, the null hypothesis is rejected which implies that the different between presence of lead in soil and water vary significantly.
TABLE 5.28 Test of significant difference of Chromium
In Soil In water
Mean 0.4455 0.244
Variance 0.7032167 0.051474
Observations 6 6
Hypothesized Mean Difference 0
Df 6
t Stat 1.696003553
P(T<=t) one-tail 0.070407178 t Critical one-tail 1.943180274 P(T<=t) two-tail 0.140814356 t Critical two-tail 2.446911846
Discussion
In terms of Lead, the presence of the chemical in soil is not significantly different from that of water since the P-value of the test is greater than 0.05 which is an indication of the existence of enough evidence to accept the null hypothesis.
Using the calculated value and the table value, the calculated value is 0.0704 and the table value is 2.45. Since the calculated value is less than the table value, the null hypothesis is accepted which implies the different between presence of chromium in soil and water does not vary significantly.
Testing of Hypothesis 3
The parasitological and bacteriological parameters of heavy metals on vegetable grown around the hospital waste dump do not vary significantly.
Test of significant difference among three selected plant (garden egg leave, water melon and cucumber);
Null hypothesis: the parasitological and bacteriological parameters of heavy metals on vegetable grown around the hospital waste dump do not differ significantly.
Using Student T-Test Statistic,
Table 5.29: Parasitological and bacteriological parameters of heavy metals on vegetable grown around UNTH waste dump
Parasitological Parameters
Bacteriological Parameter
1.6 85.1
1.86 30.09
1.26 67.32
2.18 101
1.79 32.51
1.22 56.97
1.37 61.1
1.76 27.18
1.17 33.48
Using T-test
t-Test: Two-Sample Assuming Unequal Variances
Parasitological Parameters
Bacteriological Parameter
Mean 1.578888889 54.97222222
Variance 0.119936111 697.2809194
Observations 9 9
Hypothesized Mean
Difference 0
df 8
t Stat -6.065506206
P(T<=t) one-tail 0.000150354
t Critical one-tail 1.859548033
P(T<=t) two-tail 0.000300708
t Critical two-tail 2.306004133
The P-value of the test is 0.00015 which is less than 0.05. There exists enough evidence to reject the null hypothesis and conclude that there is significant variation in the presence of the parameters in UNTH.
Parasitological and bacteriological parameters of heavy metals on vegetable grown around NOHE waste dump
t-Test: Two-Sample Assuming Unequal Variances
Parasitological Parameters
Bacteriological Parameter
Mean 64.58444444 2.238888889
Variance 480.2367028 1.058511111
Observations 9 9
Hypothesized Mean
Difference 0
Df 8
t Stat 8.52552206
P(T<=t) one-tail 1.37728E-05
t Critical one-tail 1.859548033
P(T<=t) two-tail 2.75456E-05
t Critical two-tail 2.306004133
The P-value of the test is 0.0000138 which is less than 0.05. There exists enough evidence to reject the null hypothesis and conclude that there is significant variation in the presence of the parameters in NOHE.
Testing of Hypothesis four (4)
Hypothesis: the toxic waste has no significant negative effects on rats sampled within the hospital vicinity.
Statistical analyses were performed using SPSS 2013. Shapiro-Wilk‘s (S-W) tests was used to determine the normality of data and was considered statistically significant if p value was less than 0.05. Statistical analyses were carried out after data were log transformed (normalized). Student‘s T-test was used to compare distribution of metals between livers and kidneys, and differences were considered statistically significant with p value ˂ 0.05.
ANOVA analysis were based on log transformed data done to determine the distribution pattern and possible route of heavy metals exposure to rats, using SPSS statistical software 2013. The results were shown below.
Table 5.30. Mean concentrations (± SD) and ranges of heavy metals in the livers and kidneys of rats from NOHE and UNTH.
Sample (ppm) No As Cd Co Cr Cu Mn Ni Pb Fe Al
Liver Mean 8 46 2.59a 0.198 a 0.198 a 0.147 a 18.9 a 5.50 a 0.869 a 1.05 a 263 a
SD 2.81 0.321 0.207 0.0908 14.0 2.86 0.522 2.23 95.2 2.81
Maximum 0.0699 0.0213 0.0784 0.0166 7.59 2.54 0.219 0.0532 112 0.0699
Maximum 12.6 1.45 1.21 0.406 71.8 19.1 2.11 10.7 524 12.6
Kidney Mean 8 1.91 a 1.41 b 0.382 b 0.375 b 14.3a 4.09 b 2.02 b 3.97 b 117 b 1.91a
SD 2.33 3.57 0.333 0.327 4.57 2.79 1.85 8.15 41.7 2.33
Maximum 0.102 0.0039 0.0741 0.0773 9.17 1.53 0.172 0.044 64.3 0.102
Maximum 14.0 21.1 1.62 1.72 39.5 16.8 7.85 41.6 284 14.0
n: number of samples; SD: standard deviation; different letters (a and b) between groups indicates significant difference (Student’s T-Test; p ˂ 0.05).
Source: Statistical result from SPPS, 2013.
From the Table 5.30, the mean concentrations of heavy metals in livers of the selected rats in UNTH and NOHE shows statistically difference in the level of contamination of the metals. All metals measured were detected in 100% of liver and kidney samples. Tests for normality showed a significant variation (p
˂ 0.001) in metal distribution in livers and kidneys of the selected rats in the two hospitals. Distribution of Cr, Pb, Cd, Fe, Mn, Ni, Co, Al, Cobalt and Calcium between livers and kidneys differed significantly (Student‘s T-test; p ˂ 0.05) (Table 5.30).
In order to portray the health implications of these metals to human, the results were further discussed. As, Pb, Mn, Cr, Co and Cd which were classified as the most hazardous substances by Agency for Toxic Substances and Disease Registry (ATSDR), 2013) were found to have a mean concentration of As to be higher in the rats liver (2.59 ± 2.81ppm) than kidney with 1.91 ± 2.33 ppm (Table 5.30). The liver is a major target organ of As carcinogenesis according to Onwuemesi, (2013) and could be the reason for the higher As levels. As is toxic and most hazardous substance Onwuemesi, (2013) and due to its non-biodegradable nature, it could accumulate in soil, food and water, through which rats could be exposed since they pick food and water mainly from the ground. Levels of As in soils, drinking water and organs of free-range rat raised health risk concerns for both humans and animals with food and water picking being the dominant sources in most living organisms (rat) Onwuemesi, (2012).
The high levels were attributed to incineration/burning of hospital wastes, and
this could result in the production of arsenic trioxide gas which is distributed throughout the study area via air.
As shown in Table 5.30, the mean levels of Cd in the kidney of the exposed rats from the hospitals (1.42 ± 3.57) was seven times higher (p ˂ 0.05) compared to the liver (0.198 ± 0.321) of the selected rats from the hospitals. It has been observed that animals exposed to Cd usually accumulate pollutant in their kidney because of the presence of free protein-thiol groups which leads to a strong fixation of the metal (Pompe-Gotal and Crnic, 2002). Cd concentrations in blood, urine and kidney have been recognized as good indicators of exposure by Brzoska et al., (2004). Cd could increase excretion of calcium and reduce the generation of active vitamin D in kidney.
The high level of Cd in the kidney of the exposed rat from the study showed that human are at risk especially in the areas where rats serves as source of protein. It could also lead to reduction in the free protein-thiol groups which could lead to production of active vitamin D in the kidney that could anemia especially to young children and pregnant mothers. Consequently, Cd uptake and absorption in gastrointestinal gut were observed to damage the kidney, Chen et al., (2013). Supporting the above findings, bone lesions, apart from kidney damage, are the main health consequences of chronic exposure to Cd.
Osteopenia, osteomalacia and osteoporosis with pathological fractures have been reported in Cd-exposed humans by Jarup, (2002); Alfven et al., (2000) and Honda et al., (2003).
Furthermore, the levels of Pb in the rats were higher (p ˂ 0.05) in kidney (3.97 ± 8.15ppm) than liver (1.05 ± 2.23 ppm) (Table 5.30). In some studies by different scholars, high levels of Pb was found in Manihot esculenta (cassava), soils and chickens from some communities around mining areas in Nigeria, which cause health risk to residents and especially children (Akanwa, 2016 and Onwuemesi, 2012). Bortey-Sam et al., (2015d) confirmed that the levels of Pb in organs of free-range chickens tested in Tarkwa, Nigeria emanated from contamination of soil, feeds and/or water sources, and these could be the same route through which animal were exposed. The above was in line with the present study, which shows higher concentration of Pb in the liver of eight selected rats from NOHE and UNTH when exposed to experimental analysis.
This confirmed the test, that accumulation of Pb in the liver of rat could be a source of human exposure since rat can invade human house and infect the food, water and soil leading to increase in health related diseases. Moreover, some communities in Nigeria uses rats as source of protein, when hospitals were sited without putting into consideration method of waste disposal and management, it could lead to serious health challenges to such community.
Based on the above findings, hospitals should consider location, method of wastes disposal and management before siting their hospitals and government should also make sure that hospitals were sited under due process and closely monitored to ensure sustainability.