3. A fa mous s a mpling mis ta ke
That’s Truman
They only asked rich, white
people with telephones
who’d they vote for. Sadly,
they published their
mistake
4. E ven with proper sampling…beware!
“…predicting behavior on the basis of
knowledge of attitude is a very hazardous
venture.” Meaning, predicting social
behavior is often misguided. Keep that in
mind!
6. Wha t e xa c tly IS a
“ s a mple” ?
A subset of the population, selected by
either “probability” or “non-probability”
methods. If you have a “probability
sample” you simply know the likelihood
of any member of the population being
included (not necessarily that it is
“random.”
8. S imple R a ndom S a mple
1. G e t a lis t or “ s am p ling fram e ”
2. G e ne rate rand om nu m b e rs
3. S e le ct one p e rs on p e r rand om nu m b e r
9. S ys te ma tic R a ndom
S a mple
• S e le ct a rand om nu m b e r, wh ich will b e
known as k
• G e t a lis t of p e op le , or ob s e rve a flow of
p e op le (e .g., p e d e s trians on a corne r)
• S e le ct e ve ry kth p e rs on
– C are fu l th at th e re is no s ys te m atic rh yth m to
th e flow or lis t of p e op le .
– If e ve ry 4th p e rs on on th e lis t is , s ay, “ rich ”
or “ s e nior” or s om e oth e r cons is te nt
p atte rn, avoid th is m e th od
10. S tra tifie d R a ndom S a mple
1. S e p arate you r p op u lation into grou p s or
“ s trata”
2. D o e ith e r a s im p le rand om s am p le or
s ys te m atic rand om s am p le from th e re
a. N ote you m u s t know e as ily wh at th e
“ s trata” are b e fore atte m p ting th is
b. If you r s am p ling fram e is s orte d b y, s ay,
s ch ool d is trict, th e n you ’ re ab le to u s e th is
m e th od
11. Multi-s ta ge C lus te r S a mple
1. G e t a lis t of “ clu s te rs ,” e .g., b ranch e s of
a com p any
2. R and om ly s am p le clu s te rs from th at lis t
3. H ave a lis t of, s ay, 1 0 b ranch e s
4. R and om ly s am p le p e op le with in th os e
b ranch e s
a. Th is m e th od is com p le x and e xp e ns ive !
12. The C onve nie nc e S a mple
1. F ind s om e p e op le th at are e as y to find
13. The S nowba ll S a mple
1. F ind a fe w p e op le th at are re le vant to
you r top ic.
2. As k th e m to re fe r you to m ore of th e m .
14. The Q uota S a mple
1. D e te rm ine wh at th e p op u lation looks
like in te rm s of s p e cific qu alitie s .
2. C re ate “ qu otas ” b as e d on th os e
qu alitie s .
3. S e le ct p e op le for e ach qu ota.