I was recently discussing Level 1 evaluations with some other T&D professionals. One commented that perhaps Level 1 evals correlated directly with learning: more happiness implies more learning. Another said he had heard of instructors teaching little and getting good level 1 scores by "performing" (e.g. comedy, magic or something else).

Is there any solid evidence (beyond gut feelings) that level 1 scores do or do not correlate with actual participant learning?

Wish I still had the data- but when working for a Corporate Education Center for a college in the Northeast (2003)- we cross referenced data with good level 1 scores and assessment scores.

Although in some cases there were some strong correlations (person failing test and not "getting" content causing frustration with course), there were several examples of where the "performing instructor" dazzled participants and scored well, a "not as engaging instructor" did not score well, and the assessment (learning) scores were either indifferent OR in some cases, inverse (the "performance" was more important to the instructor than the learning).

I'll see if I can find the summary findings, but I suspect I no longer have the presentation.

I think there is a strong correlation when folks "cannot get" the content- where learning and satisfaction are both low, but there are times where "the experience" and "the learning" don't really correlate.

Level 1 is an evaluation of the participant's reaction. Learning may or may not have been the participant's goal. Thus, you can't be sure that a high or low evaluation score correlates to learning.

For example, you ask all managers to attend a half-day workplace harassment course. Some managers want to learn, others already know the information, and others are closed to learning. A typical level 1 evaluation would not provide data on who learned and how much they learned in the course.

Harv has a good description of a Level 1 Evaluation (Reaction). The reaction of the participants should be about the training, the instructors performance, the types of materials used in the presentation, etc. Questions about knowledge before and after can also be asked, using a Likert scale.