If you don't have any equipment out there, you're better off buying a spec than a scope. Or none at all, and faking it... That's how it was done back in the day.
Scopes are better at the time domain, looking at periodic or transient signals and very wide bandwidths. One reason you might use a scope on an RF circuit is to check the harmonic distortion of a signal. Needless to say, you need a bandwidth many times the active frequency to observe the harmonics correctly.
Scopes are not good at observing modulation, which is not a periodic phenomenon, but manifests as blurring or jumpiness or instability in an otherwise more-or-less periodic signal. Scopes are good at observing the demodulated signal, of course (assuming a time domain representation is more useful, that is).
The bandwidth of a $400 scope is 100-200MHz, but the bandwidth of a spec of the same class is over 500MHz.
Tim
Yes,& no.
If I was going to poke around inside an FM Broadcast receiver,a Spectrum Analyser would not be my instrument of choice.
The various places you will want to look for signals inside such a receiver will not be nice AC coupled 50 Ohm output impedance.
If you are unlucky you may place a DC supply point across the input of your SA,cooking the input stage.
An Oscilloscope cannot provide the depth of information that an SA can,but you can look at the local oscillator,(this level will be a fair way down with a "100MHz"scope,but visible),then the 10.7MHz IF,examine the output of the FM detector,look at the receiver audio output,& so on.
Modulation? I would agree about FM,but AM & SSB envelopes are quite easy to display with an analog 'scope,especially if you can modulate the signal with fixed tones.
Spectrum Analysers are very useful if you can look at a device made up of modules with standard input & output impedances,but that is seldom the case with domestic equipment.