MS-ACCESS find data older than 3 years - ms-access

Question

I am trying to have a push button that finds all records that are older than 3 years. I have a push button (named Expiring) on a combo box that searches a field Intro Date in table RECS. Here is what I have:
Private Sub Expiring_Click()
Call Search
End Sub
Sub Search()
Dim task As String
Me.Refresh
task = "SELECT * FROM RECS WHERE DateDiff('m', Date(), [Intro Date])" > 36
DoCmd.ApplyFilter task
End Sub
This does not work, I get a type mismatch for my task line. I simply want a push button that will only show data that is 3 years old. . .
I am super new to MS-ACCESS and SQL/VBA. This code was based off another function I created that searches if Intro Date older than a date specified in a unbounded box.

Answer 1

correct line of code
task = "SELECT * FROM RECS WHERE DateDiff('m', Date(), [Intro Date]) > 36"
You were comparing text to the number 36 the result is a boolean instead of a string

Answer 2

As stated by others, the issue with your code is that the comparison operator lies outside of the SQL statement -
task = "SELECT * FROM RECS WHERE DateDiff('m', Date(), [Intro Date])" > 36
Should be:
task = "SELECT * FROM RECS WHERE DateDiff('m', Date(), [Intro Date]) > 36"
DateDiff Considerations
When calculating temporal differences using the DateDiff function, consider that the choice of interval argument will impact the accuracy of the result.
For example, the intuitive solution might be to calculate the three-year range by supplying the DateDiff function with the yyyy interval and the two dates in question, and simply check whether the result is greater than 3:
DateDiff("yyyy", Date(), [Intro Date]) > 3
However, observe that this will omit all dates whose year components differ by 3 or less, not dates which are exactly 3 years or fewer apart - meaning that up to an extra year's worth of time may be excluded from your results:
?datediff("yyyy", #2018-01-01#, #2021-01-01#) ' 3 years apart
3
?datediff("yyyy", #2018-01-01#, #2021-12-31#) ' 3 years + 364 days apart
3
Similarly, if you opt for the month interval (m), the remainder of a month three years in future will also be excluded:
?datediff("m", #2018-01-01#, #2021-01-01#) ' 3 years apart
36
?datediff("m", #2018-01-01#, #2021-01-31#) ' 3 years + 30 days apart
36
Obviously we now descend into an infinite regression: if days are selected, the time component will be ignored, if hours are selected, the minutes component will be ignored, ad infinitum...
And so ultimately, the choice of interval depends upon the requirements of the application: if your application requires that something expire or be validated exactly 3 years to the day, then if using the DateDiff function, the day interval (d) should be used (and note that this would include leap days, and so the resulting date could be a day prior to the date 3 years earlier); if the remainder of the month is permitted, then the month interval (m) may suffice.