FIGURE E-2 Design elements that strengthen causal inferences in observational studies.
NOTE: (a) Matching. Treatment and control stores were selected from the same chain, were in the same geographic location, and were comparable in sales during the baseline (lottery game 10). Introduction of the treatment at the beginning of lottery game 11 yielded an increase in sales only in the treatment stores. (b) Nonequivalent dependent variables. Within the treatment stores, sales of lottery tickets increased substantially following the introduction of the treatment. Sales of other major categories (gasoline, cigarettes, nontaxable groceries, and taxable groceries) that would be expected to be affected by confounding factors but not the treatment did not show appreciable change. (c) Repeated pre- and posttest measurements. Treatment and control stores’ sales showed comparable trends during the 4 weeks prior to the introduction of the treatment. The level of sales in the treatment and control stores was similar prior to the introduction of the treatment, but differed substantially beginning immediately after the treatment was introduced.
SOURCE: Adapted from Reynolds and West, 1987. Reynolds, K. D., and S. G. West, Evaluation Review 11:691-714, Copyright © 1987 by SAGE Publications. Reprinted by permission of SAGE Publications.