Education data reporting tool help

Glossary of terms

Column header definitions

National Q1

Upper value of the first quartile or the 25th percentile of the benchmark group.

National Q3

Lower value of the first quartile or the 75th percentile of the benchmark group.

National Upper CI

Upper 95% confidence limit for the benchmark group

Outcome (outlier analysis only)

The outcome (including outliers) for the report group

Response rate

The percentage of eligible trainers who completed the survey. These figures were calculated once the data had been cleansed after the survey to give improved accuracy.

SD (outlier analysis only)

Standard deviation of the report group mean

Upper CI

Upper confidence limit for the report group

Glossary of terms

Benchmark

The benchmark group is the group of respondents whose scores are used to calculate the comparative mean and the quartiles for the report you are looking at. This is not the score, but the value that the score is compared against to give it meaning. For example a score of 55 out of 100 is only meaningful if you know the average score for the benchmark group is 40. You can then use the quartiles to see how many respondents gave a score within certain ranges.

To ensure that we show the survey results in context, the benchmark group changes between reports. For example, results for cardiology at a particular trust/board are only compared to trainees in UK medical posts.

By excluding less relevant groups (in this example; surgical trainees) from the benchmark calculation, the differences between scores are more likely to reflect differences between the trusts/boards, instead of the differences found between other specialties.

You can check the list of benchmark groups used for the report you're looking at by going to the 'Applied filters' box and clicking on 'See Benchmarks A'.

Confidence limit

A confidence limit is the upper or lower value of the confidence interval. In these reports, the confidence interval is calculated against the mean indicator score for the report group.

The smaller the confidence intervals (ie the closer the confidence limits are to the mean score), the more confidence you can have that the survey results would be true across a larger sample size.

Indicator

An indicator is a measurement of a concept within the survey. These are made up of several question items on the same theme such as Educational Supervision.

To see the question items that make up an indicator and the individual question item scores for the report group you are interested in, click on the relevant bar and choose Question Items from the pop up menu.

Mean

The mean is the average score (the sum of all the scores divided by the number of respondents). There are two means for each report. One is for the benchmark group, the other is for the report group).

Median

The median is the central value of the benchmark group, or on the 50th percentile. This is calculated by identifying the middle value, after sorting numerically all the scores across all responses for the benchmark group for the report that you're looking at.

n (benchmark group population)

n is the number of respondents whose score contributed to the benchmark group.

This may not always be the full population of the survey as respondents scores are only included in an indicator score if they answer at least all but one of the questions for that indicator. This also applies to the report group population (n range).

Answering 'not applicable' to a question returns a null score. This means that if a respondent answers 'not applicable' to two questions within the same indicator, their responses to the other questions within the indicator are not included in the indicator score.

However, if you navigate to the question item analysis, the scores of the respondents excluded from the overall indicator score will be included in individual question item scores where they have answered the question.

n range (report group population)

n range represents the number respondents whose score contributed to the report group in increments of five. This is affected by the filters that you apply to the indicator analysis.

Please see the entry for n above for more information on why n or n range can vary from indicator to indicator.

Outlier

An outlier is a score that falls outside the average. This includes positive and negative scores.

Outliers are defined using two criteria (both must apply):

  • Confidence interval comparison: a below outlier has an upper 95% confidence limit that is below the lower 95% confidence limit on the benchmarking group's mean. An above outlier has a lower 95% confidence limit that is above the upper 95% confidence limit on the benchmark group's mean.
  • The quartile containing the mean: for a below outlier the mean must be in the bottom quartile and for an above outlier mean must be in the top quartile.

Percentile

A percentile is a measurement of how many scores fall into a certain range. So if in a range from 0-100 the 75th percentile is at 55, you know that 25% of the scores were higher than 55.

Quartile

Quartiles are the ranges of values based on the percentiles and medians calculated on the full range of scores.

  • Q1 is the range from the lowest score recorded to the 25th percentile.
  • Q2, also known as the interquartile range, is the range from the 25th to the 75th percentile. The median is the central value of Q2.
  • Q3 is the range from the 75th percentile to the highest score recorded.

Report group

The report group for any report is the group of respondents whose scores contribute to the bar of the graph (in indicator analysis) or the row of the chart (in outlier analysis) you are looking at.

Response rate

The percentage of eligible trainers who completed the survey. These figures were calculated once the data had been cleansed after the survey to give improved accuracy.

Trainee

Doctors in training who are in GMC approved training posts on the census date are included in the survey.

This includes:

  • Foundation trainees
  • Core trainees
  • Higher specialty training, including SpR and GP trainees
  • FTSTA trainees, LAT trainees
  • Military trainees
  • Trainees in Clinical Lecturer and academic Clinical Fellowship posts
  • Out of programme training (OOPT) on an approved programme in the UK
  • Post-CCT training towards a sub-specialty
  • Trainees working for non-NHS organisations, e.g. occupational medicine, palliative medicine and pharmaceutical medicine.

Doctors who are not in training on the census date (e.g. doctors who are on maternity leave or who have been awarded their CCT) are not included in the survey.

The survey also does not include undergraduate medical students in the process of obtaining their primary medical qualification.

Non-medical public health trainees complete a separate survey.

Trainer

All named postgraduate clinical and educational supervisors in the UK are eligible to take the trainer survey.

  • Clinical supervisors oversee a trainee's clinical work throughout a placement. They give feedback during the placement and provide a review of the trainee's performance to help educational supervisors determine if the trainee progresses to the next stage of their training.
  • Educational supervisors are responsible for the overall supervision and management of a trainee's development during a placement or series of placements. They help the trainee plan their training, achieve agreed learning outcomes and bring together all relevant evidence to determine if the trainee progresses to the next stage of their training.

Although these two roles are distinct, trainers may have responsibility for both.