A Complicated Analysis

In the classical view of cold-frontal structure, the wind-shift and temperature are essentially collocated, although high frequency data sometimes shows that there is a very short lag between the onset of the wind shift and the temperature fall.  A nice example is provided in the cross-section below, produced by Shapiro (1983).

Source: Shapiro (1983)

Actual fronts can be more complex.  Today's developing cold front provides an interesting example.  In particular, a pronounced wind-shift developed over northern Utah in advance of the surface temperature gradient.  For example, the 2200 UTC MesoWest analysis shows a very nice wind shift extending northeastward from Dugway Proving Grounds across the northern Tooele and Salt Lake Valleys.   

Time series from DPG17 (along I-80 in the heart of the Salt Flats) and Hat Island over the Great Salt Lake clearly show two distinct wind shifts, the first with little or no temperature change, the 2nd accompanied by an abrupt temperature change denoting the leading edge of the cold air.

DPG17 Meteogram

Hat Island Meteogram

As noted by Adam, the first wind shift was accompanied by a drop in dewpoint and resembled a dryline at DPG17, but this was not the case at Hat Island where the dewpoint was flat lined.  The drop in dewpoint at DPG17 also suggests that the first wind shift was not produced by diabatically induced outflow.

An interesting case worthy of further investigation.