# Guide to sampling error

## Introduction

BARB viewing data are results from a sample survey, and are therefore subject to sampling variation (also known as sampling error). Published viewing figures represent an estimate of the actual size of the audience, and may be thought of as the best estimate that can be made from a range of possible values. The lower and upper limits of this range of values can be defined in a confidence interval.

Sampling variation depends on a number of factors, including sample size, weighting, survey design and whether or not data is averaged over time. The confidence limits given in the spreadsheet are the result of modelling work carried out by RSMB, and aim to account for all these factors. The spreadsheet shows the 95% confidence limits, meaning that there is roughly a 95% chance that the true audience figure lies between the lower and upper limits of the confidence interval.

Results are given at the Network, ITV, BBC and Channel 4 Area level for twenty target audiences, and are available for the following types of audience estimate:

• Single minutes or commercial breaks
• Single minutes or breaks averaged over 4 weeks
• Half-hour programmes
• Half-hour programmes averaged over 4 weeks
• Campaign ratings

## How to use the system

Please note that the spreadsheet will only function if macros are enabled. If your browser does not prompt you to enable the macros and the spreadsheet is blank when you open it you will need to save the file to your desktop – right click the link and choose ‘Save Target As’.

To calculate your confidence interval, simply select the worksheet for the confidence interval that you require and then select the audience and enter the rating that you would like confidence interval for. Read off the result against Network or the region that you need. A full explanation is provided on the worksheet called START HERE.

### Example 1: Single minute

If we have a single ad or minute in a programme with a Network ABC1 Adult rating of 1.5 and we want to calculate the 95% confidence interval then we would use the Single Minute Rating worksheet. The confidence interval is calculated to be +/- 0.5, which means that we are 95% sure that the true value is between 1 and 2. This is equivalent to the estimated audience being 424,000 and the true value being +/-135,000 of this figure, so between 289,000 – 559,000.

### Example 2: Half hour programme

We have a 30-minute programme that has an audience of 2.3 ratings among Adults 16-34 in Scotland, which is equivalent to an audience of 26,000 viewers. To calculate the confidence interval, we would use the Half hour Rating worksheet. Selecting the right audience and entering the rating of 2.3 should conclude that the confidence interval is +/- 2.5 in Scotland, or a rating between 0 and 4.8. This is equivalent to the estimated audience of 26,000 with a confidence interval between 0 and 54,000. Note that we read the interval against the Scotland region and that the confidence interval cannot be negative.

### Example 3: Average half hour programme

We are looking at the audience slot for a particular programme to see what we typically might expect an equivalent programme to achieve in that time slot. We calculate the average 30-minute audience rating over the last four weeks to make an estimate. To estimate the confidence interval around our audience estimates we would use the Half hour Rating – 4 weeks. Assuming this was a child’s show we could see that it achieved typically achieved 3 ratings against Children 4-15, which is equivalent to an audience of 281,000. The half hour estimate has a confidence interval of 3 +/- 0.7 or 2.3-3.7 which translates to audience of between 215,000 and 347,000.

### Example 4: Campaign ratings

If we have a TV plan for an advertising campaign then we may be interested to know the confidence interval for the ratings that the plan delivers. Let’s suppose that we have an initial two week burst of 100 x 16-34 TVRs per week and then week-on, week off at 60 TVRs over a further eight weeks, for a total of 440 x 16-34 ratings. Using the Campaign ratings worksheet, we would calculate that the confidence interval is +/- 46 ratings which means that we are 95% sure that the true value is between 394 and 486 ratings. Importantly the calculation only takes in to account the days that the campaign was run, so although it was run over 10 weeks, there are only 6 weeks when it was active or 42 days in total.