The U-factor (sometimes called U-value) is a measure of the amount of heat loss due to conduction and convection. It is the reciprocal of the R-value, which measures thermal resistance (i.e., U=1÷R), so a window with a high U-factor has a low R-value. The U-factor is preferred when referring to windows. Where U-factor is reported in non-metric units of BTU/ (h-ft.2-°F), multiply by 5.678 to convert to metric units of W/(m2-°C).
A quick examination of the units reveals the usefulness of the U-factor. Whether in metric or non-metric, the U-factor measures a rate of heat transfer per unit area, per unit temperature difference. Thus, multiplying the U-factor by the window area and the appropriate temperature difference (either the average value, to get “typical” results, or the largest temperature difference for the location of interest, to get the worst case) gives a rate of heat loss, in watts or BTU/hr. This rate of heat loss (in winter) must be counteracted by supplying heat at the same rate from some auxiliary source, such as a furnace or unit heater if the room temperature is to be maintained.

Be cautious, says Marty Lapedus.

Window manufacturers may have historically quoted R-values that were for centre-glass and now refer to the centreglass U-factor when describing their product. This does not include heat loss through the frame and sash assembly, which can be significant. It is more appropriate to use total-product U-factor.
U-factors and R-values are determined either by physical tests (using ASTM or AAMA tests) or by computer simulation using the FRAME and VISION programs, as defined in the CSA A440.2 Standard.
VISION evaluates the centre-glass portion of the window, whereas FRAME analyzes the frame–sash assembly and the edge-glass region (the glazed area within 63.5 mm [2.5 in.] of the sightline of the window). In the U.S., computer-simulated values are generated using programs called THERM and WINDOW. These component values are area-weighted to define a total-product U-factor for the window, door or other envelope area of interest.
Three sources for total-window U-values are
1. the Canadian Window and Door Manufacturer’s Association (CWDMA) Certified Products List (first edition, January, 1995)
2. the ASHRAE Handbook of Fundamentals
3. the (U.S.) National Fenestration Rating Council Certified Products Directory.
Data for all sources was generated by computer simulation (using FRAME and VISION in Canada, and THERM and WINDOW in the U.S.). The CWDMA source also lists data for window SHGC, ABC ratings (resistance to air leakage, wind-driven rain, and wind deflection), energy ratings and, in some cases, condensation resistance (see the following sections).

Solar Heat Gain Coefficient

The solar heat gain coefficient (SHGC) is the amount of solar radiation incident on the exterior surface of a window that is transmitted through the window to appear as solar gains in the building. It is a decimal fraction with a value between 0.0 and 1.0 (that is, from completely opaque to completely transparent). Practical considerations limit the range of SHGC to 0.20 for a glazing system with reflective and tinted glass, to 0.87 for a single-glazed system.
Older texts refer to the shading coefficient (SC), which is the SHGC of a window relative to the SHGC of a single-glazed window at the same conditions. The SC multiplied by 0.87 provides a reasonably close approximation of the SHGC for most glazing systems.
The total-product SHGC should be used when referring to windows. Total SHGC includes solar heat gain through the frame and sash (admittedly, a small value) and gain through the glazing system, and is a smaller value than the centre-glass SHGC.
The “Fenestration” chapter of the ASHRAE Handbook of Fundamentals (chapter 31 in the 2005 edition) is a good source for SHGC values. SHGC should not be confused with visible transmission, which is discussed later in this article.