# Binomial probability calculation tool for food exposures

(Excel format, 10 sheets, 354KB)

HOW THIS TOOL WORKS:

This Excel document calculates binomial probabilities using Foodbook values as a reference population (Canada, 2015) and flags exposures of interest* for further follow-up.  There are over 200 pre-filled food items with the ability to add more items as needed. When adding additional exposures, do not insert rows. Simply add them to the bottom of the current list in order to maintain formulas and formatting.

HOW TO USE THIS TOOL:

Enter the total number of cases that reported Yes (Y), Probably (P), No (N) or Don’t Know (DK) in columns C, D, E and F, respectively, to each of the food exposures listed. The tool is then designed to do everything else using the formulas! The formulas used for each calculated cell are as follows:

• Column J calculates the percentage of cases who reported Yes (Y) and Probably (P) by dividing the sum of column C (Yes) and column D (Probably) by the sum of columns C (Yes), D (Probably), and E (No), and multiplying it my 100 to obtain a percentage.
• Column G calculates the sum of column C (Yes) and column D (Probably)
• Column H calculates the sum of  columns C (Yes), D (Probably), and E (No)
• Column K is the data from the Public Health Agency of Canada’s Foodbook study.
• Column L calculates the binomial probability of each food item using the reference data.
• Column M determines whether each exposure of interest is significant: if the p-value is less than, or equal to 0.05 AND the proportion of individuals who reported consuming a food item is greater than the Reference, OR, if no reference value exists, than if the value is greater than or equal to 60%

TO NOTE:

• The characters “#DIV/0!” appear in the table below when a number is divided by zero (0). It happens when you enter a formula like =21/0, or when a formula refers to a cell that has 0 or is blank.

Exposures of Interest are:

• those with p-values ≤ 0.05 and where the observed proportion is greater than the expected (reference) population [i.e. %Y+P > %Reference],
• OR, if there is no expected or reference value, then when the observed proportion (%Y+P) is greater than or equal to 60%

## Control Data Reference:

BINOMIAL PROBABILITY CALCULATION TABLE: