Berry's research is based on earlier research by Barry, Child, and Bacon (1959) which suggested that hunting and fishing societies tend to be individualistic, whereas agricultural societies tend to be collectivistic.  In actuality, the word "collectivistic" was not used because the research was done prior to Hofstede's famous theory of cultural dimensions. 

Berry's research had the aim of measuring the level of conformity in these two types of societies by applying a version of the Asch Paradigm.

For this sample, Berry used three distinctly different cultures.  First, the Temne of Sierra Leone, a society that is based on rice farming.  And then the Inuit people of Baffin Island in McMurdo, which survives by hunting and fishing. Each group was made up of people who had never had a western education and maintained the traditional way of life - and people who were "in transition" - either having a Western education or Western employment.  Lastly, he used Scots as a reference group.  The control group was made up of both urban and rural Scots. There were approximately 120 participants in each group. The breakdown was as follows:

Temne: Traditional (90) and Transitional (32)

Inuits: Traditional (91) and Transitional (31)

Scots: Rural (62) and Urban (60)

Each individual was brought into a room by him or herself.  For the test, they were given a set of nine lines (see diagram below).

For the first two tests, they were asked to match the line below that most closely matched the line on the top.  These were the two practice tests to make sure that the directions were understood.  The directions were given in their own language by a native interpreter using a pre-translated set of instructions and questions.

After the first two trials, there were four more trials.  On the third trial, they heard the following directions:

Here is another sheet with the nine lines on it, one here on the top and eight beneath it.  This time I am going to give you a hint.  Most Temne (or Inuit or Scottish) people say this line (experimenter pointed to a designated line) is equal in length to the one at the top.  Which one do you say?

For the third trial, the correct answer was given.  For trials 4 - 6, the wrong response was given, with each response being five lines away from the correct response. The DV was the number of lines that they were away from the correct line.  The total score could range from 0 to 15.

The results were as follows:

As you can see, the Temne, which is the collectivistic culture, had a much higher rate of conformity when told what other Temne believed, even though it was incorrect.  The Inuits, on the other hand, had an even lower rate of conformity than the Scots.

Interestingly, there was no significant difference within groups - that means, it made no difference whether the participants were living the traditional life or were highly exposed to Western culture.

As with the original Asch paradigm, the study lacks ecological validity. The task is highly artificial.

Thhe study was a quasi-experiment. Although the researchers manipulated the IV (whether they were given false information about their culture's consensus on the answer), they could not be randomly allocated to conditions.  This means that causality cannot be determined. However, the researchers used a control condition to help strengthen the internal validity of the study.  In addition, they administered the test in the native languages of the different groups to make sure that language would not be a confounding variable.

The study is an etic study, using a version of the standardized Asch paradigm.  The procedure is highly replicable and thus can be used to establish the reliability of the findings. 

The study is rather dated and so there are questions of its temporal validity. In a more globalized world with more access to media, it is questionable whether these conformity rates still are valid today. It is also important that we don't make the ecological fallacy, believing that since an individual is a member of one of these groups, that he would be more or less likely to conform.  The results of such research can lead to stereotyping about cultural groups.