Calculation of full-day equivalent working days lost estimates and rates
Estimating working days lost due to work-related illness and workplace injury
Individuals employed in the last 12 months reporting a work-related illness were asked how much time they took off work because of their illness (the most serious if more than one was reported) in the 12 month period prior to interview. It includes time lost due to all episodes of the illness in the 12 month period. Responses were assigned, by the interviewer, to one of ten categories of days, weeks or months (See Annex 2). Within each category, a uniform distribution of days off has been assumed, and individuals have been randomly assigned a number of days off according to this distribution. This new methodology was introduced in October 2015; prior to this a value was ascribed to each of these groups that represented an estimated average number of days off for the category. This was achieved by exploring other related information and earlier surveys. However, such information had become quite limited and out-of-date. Using the new method provides a more consistent approach over time, and revisions made to the back series (2000/01-2013/14) were fairly small.
Individuals employed in the last 12 months reporting a workplace injury, were asked how soon they were able to return to work after the accident. This information was reported in actual elapsed days (or weeks/months where the respondent couldn't recall the number of days and converted with the use of a standard coding frame to number of days) between injury and returning to work and does not include any subsequent time taken off work.
Missing days lost information for cases of work-related illness and workplace injuries are imputed. This is achieved by assigning a valid response from similar respondents.
Converting reported working days lost into full-day equivalent working days lost
Estimates of working days lost are expressed in the form of full-day equivalent (FDE) working days to take account of variation in the daily hours worked (for example part-timers who may work a shorter day or people who work particularly long hours). Full-day equivalent working days are calculated by adjusting the days lost estimates using the ratio of the individual's usual weekly hours to the average usual weekly hours of all full-time workers estimated using the LFS.
For this purpose, hours of work excluded overtime and meal breaks in line with the New Earnings Survey definition of full-time/part-time. Furthermore, the LFS only collects information on hours of work for current workers. Therefore usual hours are imputed for people who worked in the last 12 months but who were not working at the time of interview. Usual hours are also imputed for those current workers who did not answer the relevant usual hours of work question. Both imputations are achieved by assigning a valid response from similar respondents.
Average hours worked per week by full-time workers is based on 2001/02 data and remains constant in later surveys, allowing trends over time to be measured which takes account of changes in usual hours worked.
| Estimated annual FDE working days lost |
= | Σ [ | usual hours worked per week
Average hours usually worked per week by full-time worker |
x working days lost] |
Converting the number of workers into the number of full-time equivalent workers
Rates of working days lost are expressed in terms of average annual working days lost (full-day equivalent) per full-time equivalent worker to take into account the difference in time worked by full and part-time workers. Full-time equivalent workers are calculated as the sum of the ratio of individuals usual weekly hours to the average usual weekly hours of all full-time workers estimated using the LFS for current workers ie those working in the LFS reference week. Again the average full-time usual hours has been based on 2001/02 data (at 40.6 hours) and remains constant in later surveys.
| FTE = Σ | (usual hours worked per week)
average hours usually worked per week by full-time worker |