... a sampling bias could have potentially have had a large
impact on the results of this study, the excuses I heard not to participate had to do with
behavioral breakdowns as often as they had to do with too many after school
commitments and other measures of success ...
So basically out of 190 families scheduled to graduate, only 125 families graduated (65 did not). So right there, there are 65 failures. Out of those 125 who graduated, 25 could not be contacted. That leaves us with 25 unknown, 65 failures. Out of the 100 left, only 30 responded. Those 30 reported well due to sampling bias. That leaves 30 successes, 25 unknown, 65 failures. Out of the 70 remaining, half were still having "issues". The other half I'll give you generously as "successes". That leaves 65 successes (only 30 of which responded to a survey), 25 unknown, and
100 failures. (out of 190 scheduled to graduate). Hardly stellar, even if the numbers are accurate.