What is predicted Best Companies Index (pBCI)?

One item per factor is used to predict the pBCI. These items are:

• I am paid fairly for the work I do relative to people in similar positions in similar organisations

• I think this organisation should put more back into the local community

• I am inspired by the person leading this organisation

• I would leave tomorrow if I had another job

• My manager cares about how satisfied I am in my job

• People in our team don’t care much for each other

• This job is good for my own personal growth

• I am under too much pressure at work to perform well

Each respondent's answers to the 8 items are fed into a statistical model which predicts the values that a respondent would have selected on the BCI survey. The model was trained on a large reference dataset that learns the relationship between all the BCI items and uses this to predict BCI.

We don’t predict factor scores from this instrument, because the pBCI only includes one item per factor. We can predict an overall BCI but drilling down to individual factors from a single item each would be less reliable.

The accuracy of the prediction depends on the number of responses within an organisation or employment group. When surveying the whole organisation, the pBCI would differ from the BCI by fewer than 15.5 points in 95% of cases. At employment group level, accuracy varies with group size: with 8 respondents, the pBCI would differ by fewer than 13.6 points in 95% of cases; with 5 respondents, by fewer than 27.2 points.

We would recommend that organisations complete a full BCI survey before opting for the pBCI survey. A full survey provides the only eligible route for Accreditation and provides a richness of detail in factor and statement scores which cannot be obtained from the pBCI. Once an organisation has established a BCI, the pBCI can be used reliably for periodic pulse surveys to track progress, check the impact of interventions, or monitor employee experience between annual cycles.