# MSA 3rd Edition Bias Analysis - p. 86, the new numerical analysis of results

T

#### TommyT - 2003

MSA r.3 Bias Analysis

p. 86, the new numerical analysis of results, calculation 7)
I'm stuck. Can't calc the (tv,1-significance level/2).
Using the data in the example, I get this far:
Trials Reading Bias Reference value = 6.00
1 5.8 -0.2
2 5.7 -0.3
3 5.9 -0.1
4 5.9 -0.1
5 6.0 0.0
6 6.1 0.1
7 6.0 0.0
8 6.1 0.1
9 6.4 0.4
10 6.3 0.3
11 6.0 0.0
12 6.1 0.1
13 6.2 0.2
14 5.6 -0.4
15 6.0 0.0
0.0067 average error (bias); rounded to 4 decimals
3.55333 g=1, m=15, d*2=3.55333, Appendix C d2 table 0.225140924 std dev of range = R/d*2
15 n
3.872983346 sqrt n
0.05813 uncert. of bias is std dev of range/sqrt of n, rounded to 5 decimals
0.1153 t=bias/uncertainty for bias, rounded to 4 decimals
10.8 v=degrees of freedom, Appendix C d*2 table
3.47193 d2= infinity value of d*2, Appendix C d*2 table
0.112658697 (d2 x uncert of bias)/d*2)
0.975 1-signif level/2 from formula p.86 (tv,1-signif level/2); 1-(0.05/2) = 1-0.025 = 0.975
(tv,1-signif level/2) t = 0.1153, v = 10.8 ???????????
????? 95% Confidence Interval of the Bias, Lower ((d2 x uncert of bias)/d*2) x (tv,1-signif level/2)
What's wrong with me that I can't understand simple statistics ?
TommyT.

A

#### Atul Khandekar

Tommy,

No T-table would give you the T-statistic for DF=10.8. You will get T values for DF=10 (T=2.22814)and 11(T=2.20098). You will have to 'interpolate' from those values to get the significant t-value(2-tailed)= t(v,1-signif level/2) =2.2064. The value of t=0.1153 is just not used!!

-Atul

T

#### TommyT - 2003

Well,I don't understand why the t=0.1153 is bothered to be calculated, as it shown on p.86 t=bias/uncertainty of bias , and it shown on table 3 on p.88
The interpolated 2.206 (also shown as a result in Table 3) does work very well, I don't see the reason to calculate t, do you ?
Thanks.
TommyT