Evaluation

• The study relied on self-report, which means social desirability may have been a factor


  participants may have been embarrassed about some of their dreams or fantasies and so may not have reported them at all or may have missed out or changed aspects in their description. If these things occurred, it makes comparison unreliable.
• The difference in scores may be down to variabilities in reporting techniques or the setting the reports were written in


  participants were asked to report their dreams or fantasies as they happened but this may not always have been practical. This means that details may have been forgotten, or that participants took time to make sense of what they remembered giving a more coherent yet accurate description.
• There was a lack of control over the independent variable, which was supposed to be dreams versus fantasies


  because the results relied on participants reporting from home, it is possible that they were writing about dreams that took place in non-REM sleep, rather than REM sleep. In addition, some fantasies may have taken place when people were feeling drowsy and so the brain functions like it does in non-REM sleep. Due to these extraneous variables, the differences between dreams and fantasies may have actually been greater than the results suggested.
• The sample was difficult to generalise. Although a number of reports were analysed, they came from only twelve different people


  also the majority of these were females, which could affect the nature of dreams and fantasies, fiving results which were gender biased.
• The results mat lack construct validity. This is because dreams and fantasies are complex phenomena


  even if they were described accurately, the fact they were rated and reduced to numbers means that they are oversimplified. Critics argue that data on dreams should be qualitative rather than quantitative.