Expectancy effects

A topic in research methodology

One methodology (general research strategy) is the experiment. The aim is to find the effect of a changing variable (the independent variable) on a variable that is being observed/measured (the dependent variable) whilst controlling (keeping constant) and other factors that might affect the observation / measurement (control or controlled variables).

Read about experiments

Confounding variables

A confounding variable is anything that varies between the experimental and control conditions, and which might influence outcomes, but is is not controlled or allowed for.

Read about confounding variables

One such variable might be that those involved (as investigators, and/or participants) have different expectations about outcomes in the different conditions – as such expectations can influence outcomes. This is not just a hypothetical risk, but something that has been often demonstrated:

The Pygmalion effect

"This was demonstrated in a study where teachers in a school were told that tests on the children had identified those – 'growth-spurters' – who were likely to make higher levels of progress in the following school year…. These predictions came true: statistically, the identified children did significantly better in school than their classmates after their teachers had been told of their status as growth-spurters. Actually, these children had been assigned this label at random, so the results were either an unlikely chance event, or were somehow the outcome of teachers' expectations mediating class- room processes. That is, either by chance the students identified just happened to be those who were indeed going to make better than average progress in the next school year (and this is not logically ruled out by the statistics, but rather just shown to be very unlikely) or there was a substantive effect due to teachers knowing who had been identified as about to make good progress. The students that teachers expected to do well actually tended to do well even though they had been selected purely by chance. A great many other studies have since replicated effects of this kind…" (Taber 2019: 77)

Pygmalion?

In myth, Pygmalion was a sculptor who created a sculpture of a perfect woman – so perfect he fell in love with his own creation. He asked Aphrodite that he might find a bride who looked like his stature – and she granted his wish by bringing the statue to life. Perhaps atypically, for such a myth (be careful what you wish for!), they were happily married – so Pygmalion offers a paradigm case of wish fulfillment!

What is the mechanism?

Statues coming to life would be difficult to explain. More generally, however, we can understand why our expectations can indirectly have an effect.

"In medicine, patients may have a lot invested in the promise of a new drug treatment, and those receiving an experimental treatment may be looking for any small sign that the medicine is working for them – whilst those assigned to a control condition may feel disappointed, having enrolled in a trial in the hope of getting the new experimental drug. If clinicians are optimistic about the new drug, their expectations may be inadvertently communicated to patients, or may bias their measurements of effect when rating subjective reports of symptoms, for example." (Taber 2019: 75-6)

"Similar threats to validity are at work in educational settings…Researchers and teachers may be optimistic about some new teaching approach or curriculum materials and this could bias their judgements, and change their classroom behaviour. Teachers may subtly communicate their expectations to learners who may also respond to a teacher's additional enthusiasm for, and commitment to, an intervention, even if they are not directly aware that the teaching is in some way different from the norm.
This is clearly a major issue in experimental research in education…The response to such a threat used in drug trials – doing the research double blind – is seldom an option in education as it is usually obvious to researchers, teachers, and even learners, when they are part of an experimental treatment condition." (Taber 2019: 76)

Avoiding expectancy effects

Even when aware of the possibility of expectancy effects, it may be difficult to avoid them. In drug trials,

"This is readily avoided if neither patent nor doctor knows who is getting which treatment (a situation known as double-blind), and the analysts are working with anonymised data…The response to such a threat used in drug trials – doing the research double blind – is seldom an option in education as it is usually obvious to researchers, teachers, and even learners, when they are part of an experimental treatment condition." (Taber 2019: 75-6)

In one study of the effect of formative assessment the researchers claimed that "students were not aware of the condition they participated in" (Vogelzang & Admiraal, 2017: 159". However, all participants had consented to (and so were aware of) the study, and if the formative assessment activities were innovative in the research context this might well have offered clues. In any case, it is the knowledge and expectations (and sometimes enthusiasm) of the researchers which is considered to often mediate outcomes shown by learners. So, we should be aware that expectancy, or indeed novelty, effects could be pertinent.

Sources cited: