Beruflich Dokumente
Kultur Dokumente
__
F-test:
F=
s1 s2
2 2
___
t = s
x1 x 2 1 1 + n1 n2
___
= 2.27
Degrees of freedom = n1+n2-2 = 5+5-2 = 8 tcritical= tn-1=t8= 2.306 at 95% confidence level Since tcritical > t, it implies that there is no statistically important difference between the means of the two methods. The presence of a small difference between the means can be attributed to random errors. Beech Leaves
__
F-test:
F=
s1 s2
2 2
s=
(n1 1) s1 + ( n2 1) s 2 = (n1 + n 2 2)
___ ___
t = s
x1 x 2 1 1 + n1 n2
= 5.26
Degrees of freedom = n1+n2-2 = 5+5-2 = 8 tcritical= tn-1=t8= 2.306 at 95% confidence level Since tcritical < t, it implies that there is a statistically significant difference between the means of the two methods and only one method can be used to determine chromium in the organic materials. Aquatic Plant
__
F-test:
Since, Fcritical > F implies that there is no significant difference between the two variances and the t-test assuming equal variances can be used to compare the two experimental means. t-test assuming equal variances:
s=
(n1 1) s1 + (n 2 1) s 2 = ( n1 + n2 2)
___
t = s
x1 x 2 1 1 + n1 n2
___
= 3.73
Degrees of freedom = n1+n2-2 = 5+5-2 = 8 tcritical= tn-1=t8= 2.306 at 95% confidence level Since tcritical < t, it implies that there is a statistically significant difference Researc her A 84.32 84.51 84.63 84.61 84.64 84.51 B 84.24 84.25 84.41 84.13 84.00 84.30 C 84.29 84.40 84.68 84.28 84.40 84.36 D 84.14 84.22 84.02 84.48 84.27 84.33 E 84.50 83.88 84.49 83.91 84.11 84.06 F 84.70 84.17 84.11 84.36 84.61 83.81 between the means of the two methods and only one method can be used to determine chromium in the organic materials. Table 1 Paracetamol Content (% m/m)
2.
Joshua Pinto- A817206 By using the single factor ANOVA test tool in excel, the results in Table 2 were obtained. Table 2: ANOVA single test excel output SS df MS F P-value 0.57182 5 1.33565 1.90747 5 5 30 35 0.11436 5 0.04452 2 2.56874 9 0.04756 2
F crit 2.53355 5
From Table 3, F > Fcrit which implies that the percentage of paracetamol obtained from the tablets differs significantly with different researchers and the null hypothesis is rejected.
To determine the cause of this difference, the least significant difference method was used. Table 3: Anova: Single Factor Count Sum Avera ge 6 507.22 84.53 7 6 505.33 84.22 2 6 506.41 84.40 2 6 505.46 84.24 3 6 504.95 84.15 8 6 505.76 84.29 3
Researcher A B C D E F
2 t h ( n 1) n
2 2.042 = 0.249 6
Table 4 illustrates the means of researchers arranged in increasing order according to results from Table 3;
Table 4: Increasing order of means Researcher Mean E 84.158 B 84.222 D 84.243 F 84.293 C 84.402 A 84.537 Comparing the mean values in table 4 to the least significant difference, there is a significant difference in the paracetamol content obtained by researcher A (> 0.249) compared to the paracetamol content obtained by researchers B, D and E.
Sample No
Anitomony Found (mg m-3) New Method 22.2 19.2 15.7 20.4 19.6 15.7 Standard Method 25.0 19.5 16.6 21.3 20.7 16.8
1 2 3 4 5 6 3.
To determine if the two methods differ significantly, the t test for paired two sample of means has been carried out using excel and is shown in table 4. Table 6: t-Test -Paired Two Sample for Means New method 18.8 6.828 6 0.97327052 5 0 5 3.43111296 8 0.00930657 5 2.01504837 2 0.01861315 2.57058183 5 Standard method 19.98333333 9.845666667 6
Mean Variance Observations Pearson Correlation Hypothesized Mean Difference df t Stat P(T<=t) one-tail t Critical one-tail P(T<=t) two-tail t Critical two-tail
Table 7: Statistical values of G for suspected values of b Suspect Value Mean Standard Statistic G of b deviation, s 0.4000 0.404 0.0045 0.7762 0.3900 0.401 0.0083 1.3275 0.3800 0.399 0.0129 1.4316 0.3700 0.396 0.0178 1.4642 __ 0.3600 value 0.394 0.0227 1.4781 suspect x 0.3570 0.393 0.0241 1.4807 Statistic G = s 0.3571 0.393 0.0241 1.4806 As shown in Table 5, l t l > tcritical which means that the two methods provide significantly different amounts of Antimony from the atmosphere.
4. 0.403, 0.410, 0.401, b To test possible values of b before it is rejected as an outlier, the Grubbs test has been used.
Gcritical for a two-tailed test with sample size of 4 = 1.48 (P=0.05) Therefore, comparing the critical value of G to the statistical G values from table 5, the lowest possible value of b before it is rejected as an outlier is approximately 0.357 (3 d.p.).
Joshua Pinto- A817206 Another test that can be used to determine the lowest value of b before being considered an outlier is Dixons test (Q-test) and is shown below;
Statistic Q= suspect value nearest value value
Table 8: Statistic values of Q for suspected values of b Suspect value of b Statistic Q 0.4000 0.1000 0.3900 0.5500 0.3910 0.5263 0.3920 0.5000 0.3930 0.4706 0.3800 0.7000 0.3700 0.7750 0.3600 0.8200 0.3571 0.8299 Qcritical for a two-tailed test with sample size of 4 = 0.829 (P=0.05) Therefore, comparing the critical value of Q to the statistical Q values from table 6, the lowest possible value of b before it is rejected as an outlier is approximately 0.357 (3 d.p.) The results from the Grubbs test and Dixons test conclude the same results i.e. the lowest possible value of b before it is rejected as an outlier is approximately 0.357.
5. a)
Concentration, 10 -1 ng ml Using the data from table 9, a graph was plotted as shown in Figure 1 to determine the equation of the line
From Figure 1, the equation of the line was determined to be y=0.025x + 0.002. Using the regression analysis tool in excel, table 10 was produced as shown below from which the 95% confidence limits of the slope and intercept were obtained
Joshua Pinto- A817206 Table 10: Regression output t Stat P-value Lower 95%
Standard Error
Upper 95%
0.00478739 3 0.00026555 7
0.440144 11 94.76046 84
0.0101992 43 0.0244816 5
0.01441352 9 0.02584692 1
-0.010199243 0.02448165
Upper 95.0% 1.3751E307 28.9190456 4 8.8862E+1 84 8.8862E+1 84 3.4569E+1 84 1.2645E289 0.01441352 9 0.02584692 1
Intercept (a) = 0.0021 0.0123 Slope(b) = 0.0252 0.0007 Error in the intercept: 0.002107143 0.010199243) =0.01231 0.0123 or 0.014413529 ( 0.002107143= 0.01231 0.0123 Error in the slope: 0.025164286 0.02448165 = 0.000683 0.0007 or 0.025846921 0.025164286= 0.000683 0.0007
10
Joshua Pinto- A817206 b) In order to estimate the 95% confidence limits for the silver concentration in a sample with an absorbance value of 0.456, the standard deviation of the concentration was first calculated which is given by;
__
sx 0 =
sy/ x b
1 1+ + n
( y0 y ) 2 b 2 ( xi x ) 2
i __
Where, y 0 = 0.456
11
The values of other parameters were obtained from table 11 and 12; Table 11: Summary output of Regression Regression Statistics Multiple R 0.999721706 R Square 0.999443489 Adjusted R Square -1.4 Standard Error 0.007025972 Observations 1 Table 12: Regression calculations xi 0 5 10 15 20 25 30 105 15 xi2 0 25 100 225 400 625 900 2275 y 0.003 0.127 0.251 0.390 0.498 0.625 0.763 2.657 0.380 x- x -15 -10 -5 0 5 10 15 0 =r = Sy/x
Mea n
s x 0 =
12
Joshua Pinto- A817206 Thus, the confidence limits of the concentration are given by;
x 0 t ( n 2 ) s x0 t ( n 2 ) = t 5 = 2.571 ( P = 0.05 ) x0 = y 0 a 0.456 0.002107 = = 18 .04 ng / ml b 0.02516
Therefore, the 95% confidence limits for the silver concentration in a sample with an absorbance value of 0.456 is
1 .0 ( 2.5 1 0.3 0 ) 8 4 7 0 1 .0 0.8 n / m 8 g l
c) The minimum concentration of silver that can be measured by FAAS was estimated using the limits of detection of instrumental analysis which is given by; Limit of detection = a + 3sy/x = 0.0021+ (3 x 0.00703) = 0.0232
x0 = 0.0232 0.0021 = 0.837 0.84 ng / ml 0.0252
Therefore, the minimum concentration that can be measured by FAAS is 0.84 ng/ml
d) In order for FAAS to analyse a sample in which the silver concentration is below the detection limit, an alternative wavelength with a higher absorptivity can be used or the path length can be increased by rotating the burner hand of the instrument which would enable the detection of the silver concentration within the linearity response range of the instrument. On-line solid phase extraction (SPE) may also be used as an alternative to flow injection liquidliquid extraction in order to improve the sensitivity and selectivity of FAAS to detect silver concentrations below the detection limit.
6. Table 13 t/s CT (kmol/m 3 ) 0 240 00 100 23100 300 2200 0 500 2090 0 700 1900 0 1000 1730 0
13
Joshua Pinto- A817206 Using the data from table 13, a graph of oxygen concentration in the reservoir vs. Time was plotted as shown in figure 2.
The equation of the line was found to be y= -6.686 x + 23948 The gradient of the graph represents dCT/dt = -6.686
rA =
The negative sign in the rate of reaction implies that A is being consumed in the reaction. Using the regression function in excel, the results in Table 14 were obtained Table 14: SUMMARY OUTPUT Regression Statistics Multiple R 0.99690420 =r 9 R Square 0.99381800 2 Adjusted R -1.5 Square Standard Error 222.716869 =S To calculate the 95% confidence limits of the rate of reaction, the standard deviation of slope sb = was first calculated. This is given by Observations 1 14
y/x
sy / x
( x
i
x )2
__
= 222.72 Table 15: Regression calculations x 0 100 300 500 700 1000 2600 433.33 x2 0 10000 90000 250000 490000 1000000 1840000 x - x mean -433 -333 -133 67 267 567 0 (x - x 2 mean) 187778 111111 17778 4444 71111 321111 713333
Mea n
(x
i
__
s b =
= 0.264
t ( n 2 ) = t 4 = 2.776 ( P = 0.05 )
Therefore, the 95% confidence limit of dCT/dt = 2.776 0.264 = 0.732 The rate of reaction is related to dCT/dt by a multiplicative expression. Therefore the 95% confidence limit of the rate of reaction can be calculated by the formula:
2
rA
dCT = dt dC T dt
15