Trump was using the percentage of non-participation in the work force as a measure of unemployment.
I just got hold of a Merrill Lynch publication that speaks to this.
Workforce participation has steadily grown since the 1960s driven by the advent of the booms and women going to work.
It increased during the 1980s and reach a peak in 1999 at the top of the DotBomb bubble when it was right at 67% (33% unemployment according to the Donald).
It took a dip with the bursting of the DotBomb and Enron bubble then flattened out to 66% in 2009.
Since 2009, it has made a straight line decline to 62.7%.
There was a second graph which showed by age group changes from 1992 to 2012:
75+ went from 4% to 9%
65-74 went from 17% to 28%
55-64 went from 57% to 64%
25-54 went from 83% to 81%
20-24 went from 78% to 71%
Note the precision of these numbers may be a little weak because I was reading off a 2" x 1-1/2" picture.
The numbers purprotedly come from BLS in Oct 2015.
Looks like people are working longer and squeezing out the youngest workers.