Detection Limits—There Is No Such Thing as Zero

Cell Cl 1 contains the standard error of the estimate (or standard deviation of the regression) and is a measure of the error in estimating values of y. The smaller it is, the closer the numbers are to the line. The other cells contain data we will not consider here Cell B12 is the F value, cell C12 the degrees of freedoin (used for F), cell B13 the sum of squares of the regression, and cell C13 the sum of squares of the residuals. 

How many significant figures should we keep for the least-squares line The standard deviations give us the answer. The slope has a standard deviation of 1.0, and so we write the slope as 53.8 l.o at best. The intercept standard deviation is 0.42, so for the slope we write 0.6 0.4. See also Example 3.22. 

Excel offers a number of statistical functions, listed under the Tools menu. Go to Add-Ins, and check Analysis ToolPak. Click OK and return to the spreadsheet. Now when you go to the Tools menu, you will see Data Analysis. Go to that, and you will see 19 statistical programs listed. As you experiment with these, you will find some very useful. One Add-In that is very useful is Solver, for solving complicated formulas. Its use is described in Chapter 6. See also the text website www.wiley. com/college/christian for a list of some commercial software packages for performing basic as well as more advanced statistical calculations. 

22 Deletion Limils—There Is No Such Thing as Zero 

The previous discussions have dealt with statistical methods to estimate the reliability of analyses at specific confidence levels, these being ultimately determined 

---

DETECTION LIMITS—THERE IS NO SUCH THING AS ZERO... 

---