Calculating a Bradford score from a list of dates.

Given the revised data (i.e. all expected days included) a single-cell {Frequency} calc, per Associate, will be simpler -- however I confess I'm pushed for time so, using sample file:

Formula:

D2: =IF($A2<>$A1,1,N(D1)+($C2<>$C1)*$C2)

E2: =IF($A2=$A3,"",($D2^2)*COUNTIFS($A$2:$A2,$A2,$C$2:$C2,TRUE))

copied down

the above would generate your expected results, I think... if you want to make more elegant post back -- alternatively one of the other contributors can provide for you.

Hope that helps.

edit: in answer to your question...

The 1 in O2 is that the employee number? The '0001011 if P2 does that represent Thursday some how?

Yes, the 1 was to denote the ID... the 7 digit string (prefixed with apostrophe) is used to record binary value for Mon-Sun, where a 0 represents a working day
{this allows NETWORKDAYS.INTL to calculate working days between two dates relative to any work week permutation}

the above isn't required given revised data-set.