Some studies can avoid these problems by having participant respond in writing(and then having their responses typed and scored by raters who do not know the gender of the writer),but this limits the kinds of observations and data that can be collected.