John Gilmer wrote:

... A 10" diameter globe would only take a 15 watt CF. In theory
that's as bright as a 60 watt conventional light but who is kidding!

It's close. You can measure the relative outputs with a Bunsen grease
spot photometer. Put a bulb at each end of a yardstick and move a piece
of white paper with a grease spot between them until the spot disappears,
indicating equal illumination from both sides, at which point I1/d1^2
= I2/d2^2 for distances d1 and d2 from bulbs 1 and 2.

Thank you.

You are welcome.

I might try that some day. You method "works", I guess, because the spot
causes some light to be transmitted through that otherwise would be
reflected. When the light from the back side makes up for the light
transmitted the spot goes away.

The method comes from Robert Bunsen (1811) of burner fame.

Maybe I can get my older girl to try it as a "project".

A science fair project might have a meter stick with each bulb
screwed into a 2-prong socket plugged into a Kill-A-Watt meter.

The KAW showed a 0.57 power factor for a 14 watt Commercial Electric CF from
a recent Home Depot 6-pack, vs 1 for the incandescent. For extra credit, she
might try correcting the CF's power factor to 1 with a small parallel motor
run cap, based on the measured PF, with reactance Xc = V^2PF/(Psqrt(1-PF^2))
and capacitance C = 1/(377Xc), ie 2653Psqrt(1-PF^2)/(V^2PF) microfarads, eg
C = 2653x14sqrt(1-0.57^2)/(120^2x0.57) = 3.7 uF. Motor repair shops often
discard used motor run caps.

Will this work for an electronic vs magnetic ballast?

Nick