The estimated proportional area occupied by each site series within a TSA or TFL is calculated in the data analysis phase of the project. The statistical data analysis approach was developed for the Ministry of Forests (Linnell Nemec 1998) and its details are included in Appendix 2. In addition to the proportional area calculations, this report also presents information pertaining to sample size.

After the data collection phase is complete, the plot data should be entered and compiled in the Excel spreadsheet, including the biogeoclimatic zone, subzone/variant, site series, and proportion of the plot within that site series. An example of a portion of the spreadsheet compiled for the Bulkley TSA is provided in Exhibit 3.2.1. Using Linnell Nemec (1998) for guidance, calculate the proportional area and its standard error for each site series within the entire TSA.

1. The spreadsheet is designed so that each line in the spreadsheet represents one site series in a plot. Therefore, if a plot covers three site series, then it will have three lines in the spreadsheet. Enter the proportion of the site series within the plot, e.g., if the site series covers 200 m², then the proportion is 200 ¸ 400 = 0.5.
2. Sort the Excel data file by biogeoclimatic zone, subzone/variant, and site series.
3. Calculate the proportion of the site series in the TSA/TFL.
1. Sum the proportions estimated for each site series from all the samples.
2. Divide the sums calculated in step 3 by the total number of all sample plots (call it n, e.g., n = 500). This gives the proportion of the total TSA/TFL occupied by the site series.
4. Calculate the standard error of the estimated proportion.
1. Calculate the square of the proportion for each site series within each plot in the spreadsheet.
2. Sum the squared proportions for each site series.
3. Square the value obtained in step 3a and divide this by n.
4. Subtract the value obtained in step 4c from the value obtained in step 4b.
5. Divide the value obtained in step 4d by n ´ (n – 1).
6. Take the square root of the value obtained in step e. This will be the standard error of the proportion calculated in step 3.
5. These proportions can be also be calculated by zone or subzone/variant. To do this, just change the number of samples (n) to the number of samples in the zone or subzone/variant of interest in the above calculations.

Exhibit 3.2.1 shows examples of these calculations. Note that in this Exhibit (an Excel spreadsheet) the above calculations are programmed into the spreadsheet.

The standard errors are not used directly in the timber supply analyses. However, 95% confidence intervals can be constructed with the standard errors by multiplying them by 1.96 and adding and subtracting them from the proportions. This will give an indication of the accuracy of the estimated proportion. If the confidence interval is too wide, additional samples could be established to improve the accuracy of the estimated proportions. What constitutes too wide is a subjective decision which depends at least partly on acceptable risk. Timber supply branch should be contacted if you have questions about accuracy and acceptable risk.

Continue to Chapter 6.
Return to Table of Contents.