Tipultech logo

Stubs about research methods

Author: Dr Simon Moss


Killeen, P. R. (2005). An alternative to null-hypothesis significance tests. Psychological Science, 16, 345-353.


Syptoms of autism are now assumed to permeate the general population to various extents. Indeed, Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley (2001) developed a scale to assess traits associated with autism in adults with normal intelligence. The measure comprises three facets: deficient social skills (e.g., "I am good at social chit-chat", reverse scored), defective communication (e.g., "People often tell me I keep going on and on about the same thing"), and attention to details (e.g., "I notice patterns in things all the time.

These symptoms, according to Liss, Mailloux, and Erchull (2008) might reflect a need to withdraw because of intense stimulation--as a consequence of processing sensitivity to sensory information.


Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E. (2001). The autism-spectrum quotient: Evidnce from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of Autism and Developmental Disorders, 31, 5-17.

Liss, M., Mailloux, J., & Erchull, M. J. (2008). The relationship between sensory processing sensitivity, alexithymia, autism, depression, and anxiety. Personality and Individual Differences, 45, 255-259.


Woods, S. A., & Hampson, S. E. (2005). Measuring the Big Five with single items using a bipolar response scale.

European Journal of Personality

, 19, 373-390.


Often, researchers will report several studies with multiple, but related, dependent variables--all assessing the same hypotheses. For example, they might conduct two studies to assess whether studying statistics enhances IQ. In each study, two measures of IQ are used: WAIS and an SHL test. Hence, four p values and z values emerge.
In this instance, the tests are patently not independent, and thus common meta-analytical techniques do not apply (see Strube, 1985).

Researchers often want to ask three questions in this instances. First, overall, is the hypothesis supported. Second, does the extent to which the hypotheses are supported vary across the tests, suggesting a moderator might be operating. Third, can a moderator by established.

If the z values were independent, the best formula merely involves summing the z values and dividing by the square root of the number of tests Strube provides a formula to accommodate the situation in which some, or all, of the z values are not independent--that is, if some of the z values are derived from the same individuals. Specifically, the denominator becomes

Sum of the variances plus twice the sum of the covariances. The variances are one--because they are standard scores. The covariances are zero for independent hypotheses. The covariances become the correlations for dependent hypotheses. This test becomes more conservative.


Strube, M. J. (1985). Combining and comparing significance levels from nonindependent hypothesis tests. Psychological Bulletin, 97, 334-341.


Self efficacy measures might be more effective if individuals estimate the confidence of others, not themselves, to prevent impression management and self deception (e.g., Demmel, Nicolai, & Jenko, 2006)


Demmel, R., Nicolai, J., & Jenko, D. M. (2006). Self-efficacy and alcohol relapse: Concurrent validity of confidence measures, self-other discrepancies, and prediction of treatment outcome.

Journal of Studies on Alcohol

, 67, 637-641.


Confidence to seek participants

Many researchers feel uneasy when they approach potential participants. To overcome this unease, as well as to reduce the likelihood of rejections, researchers should first imagine they were approached with the same request. That is, they should imagine themselves from the perspective of someone whose assistance is sought. Second, they should recognize the shame or embarrassment they would feel if they refused to help. Third, they should acknowledge to themselves the other person is likely to experience the same shame or embarrassment--and, consequently, will often feel obliged to help. Fourth, researchers should directly highlight their need to seek help, with explicit statements such as "I need some help...".

These questions merely exacerbate the shame that individuals experience when they refuse to help (Flynn & Lake, 2008). Indeed, Flynn and Lake (2008) showed that individuals are twice as likely to comply with requests to participate in studies than researchers anticipate.


Flynn, F. J., & Lake, V. K. B. (2008). If you need help, just ask: Underestimating compliance with direct requests for help. Journal of Personality and Social Psychology, 95, 128-143.

Academic Scholar?
Join our team of writers.
Write a new opinion article,
a new Psyhclopedia article review
or update a current article.
Get recognition for it.

Last Update: 5/31/2016