when visitors arrive and placing a sticker on their clothing to alert the videographer (Crowley and Callanan, 1998), or getting explicit consent as visitors enter a space. Such methods are generally compromises, and researchers should always refer to their local institutional review board for approval of their specific data collection method.
In environments such as museums, botanical gardens, and zoos, where learners move freely through a physical space of options, time spent (“holding time” or “dwell time”) is a commonly used measure of impact in summative evaluations. At the same time, there is controversy about what exactly it assesses in relation to learning. There are various approaches to thinking about time, including:
Some researchers regard it as a necessary but not sufficient condition for learning. In this view, learners need to pause and engage with objects, people, or activities in order to have a chance to learn from them, but learning is not necessarily linearly related to time spent. Some researchers have interpreted histograms of holding time as bimodal or multimodal, revealing different audience characteristics in terms of background or motivation (browsers, grazers, etc.), but these are controversial: most exhibitions show a single peak at the short end of the spectrum of time spent (Serrell, 1998, 2001).
Some regard it as an indicator of learning, using the well-established principle that time on task is the most universal correlate with learning across contexts. However, the meaning of “on task” is particularly ambiguous in free-choice environments (Shettel, 1997), as is the definition of learning. A few studies have shown direct evidence that time spent in exhibitions correlates with learning, as measured by previsit questionnaires on the exhibit topic (Abler, 1968) or free recall of objects seen (Barnard and Loomis, 1994).
Some regard time spent as a direct measure of learning, defined as engagement in socially sanctioned collaborative activity. From this socio-cultural perspective, participants are learning throughout their engagement, although the exact nature of what they learn may be quite different from institutional expectations.
Increasingly, the Internet is being used to conduct surveys of learners. These may be assessments of online resources or may ask about previous experiences in another setting (such as a museum visit, viewing of a TV series, etc.). They may be contained within emails, or, increasingly, be web-