The internal, external, and construct validity are all affected by the use of a quasi-experimental design, and play an important role in assessing the program outcomes. Internal validity of a program is the extent to which a causal conclusion is warranted, and this is “determined by the comparability of its experimental and control groups” (text 186). To address internal validity a comparison group must be created, which it was in the case of OSGV. The comparison group is intended to be “so similar to the experimental group that it will show what the experimental group would look like if it did not receive the experimental treatment” (186 text). The comparison groups of OSGV were not almost identical to the experimental as they should have been to aid in strengthening the internal validity of the program. Rather the comparison groups were 20 blocks that had gang rates “not as high” as the 40 experimental blocks. While creating comparison groups strengthens internal validity, improperly choosing the comparison groups weakens internal
The internal, external, and construct validity are all affected by the use of a quasi-experimental design, and play an important role in assessing the program outcomes. Internal validity of a program is the extent to which a causal conclusion is warranted, and this is “determined by the comparability of its experimental and control groups” (text 186). To address internal validity a comparison group must be created, which it was in the case of OSGV. The comparison group is intended to be “so similar to the experimental group that it will show what the experimental group would look like if it did not receive the experimental treatment” (186 text). The comparison groups of OSGV were not almost identical to the experimental as they should have been to aid in strengthening the internal validity of the program. Rather the comparison groups were 20 blocks that had gang rates “not as high” as the 40 experimental blocks. While creating comparison groups strengthens internal validity, improperly choosing the comparison groups weakens internal