Ethical Issues and Guidelines for Conducting Data Analysis in Psychological Research

Wasserman, Rachel M.

doi:10.1080/10508422.2012.728472

Cited by 14 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Failure to do so introduces unnecessary error variance into data, which reduces statistical power and the likelihood of detecting meaningfully sized effects (Maniaci & Rogge, 2014). It should be noted that such carelessness and/or incompetence, whilst not representing research misconduct, may still be considered unethical (Wasserman, 2013).…”

Section: Discussionmentioning

confidence: 99%

Detecting Duplication in Students’ Research Data: A Method and Illustration

2015

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Detecting Duplication in Students’ Research Data: A Method and Illustration

2015

View full text Add to dashboard Cite

“…Accordingly, it is deemed inappropriate to collect (primarily quantitative) data, run a number of statistical tests, see which hypotheses are supported by the data, and then describe the hypotheses as if they were formulated before data collection. This is sometimes referred to as a fishing expedition or as HARKing (hypothesizing after the results are known; Wasserman, 2013) and represents a clear ethical violation for many. This is the case for good reason.…”

Section: The Importance Of Developing and Changing Research Questionsmentioning

confidence: 99%

“…This is the case for good reason. Running a large number of statistical tests (especially if these are exploratory rather than confirmatory; Wasserman, 2013) increases the likelihood of type I error rates (Simmons et al, 2011;Bakker et al, 2012). The American Statistical Association states that "Running multiple tests on the same data set at the same stage of an analysis increases the chances of obtaining at least one invalid result.…”

Section: The Importance Of Developing and Changing Research Questionsmentioning

confidence: 99%

I Never Promised You a Rose Garden: When Research Questions Ought to Change

MacIntosh¹,

Bartunek²,

Bhatt³

et al. 2016

Research in Organizational Change and Development

View full text Add to dashboard Cite

This chapter addresses a topic which is rarely discussed but is a common feature of field-based action research, namely whether and when research questions can legitimately change during the conduct of a study. We explore three common strategies for developing research questions before problematizing the assumption which these strategies share, namely that research questions are fixed at the outset and remain stable. We note that change can, does and indeed should occur in response to changes in the context within which the research is being conducted. Using an illustrative example of a longitudinal project, we identify refinement and reframing as two distinct types of research question development that might occur. Our conclusions suggest that greater transparency over when, in what circumstances and why researchers elect to change their research questions would be a healthy development for the field.

show abstract

“…Another example is that the use of the Internet to conduct studies, instead of in person, may present new ethical challenges in ensuring that research par ticipants have given informed consent and have been adequately debriefed (see Pittenger, 2003). There are even ethical concerns about the use of new complex statistical techniques because of the possibility that psychologists who lack a thorough understanding may misapply them, or perpetrate other irregularities in data analysis that lead to erroneous conclusions (see Wasserman, 2013). We have highlighted only a few examples in order to demonstrate the pervasive ness of ethical concerns in contemporary psy chology, and the possibility that today's "ethics mindset" might be coloring our interpretations of the Little Albert study.…”

Section: Why This Should Matter To Historians Of Psychologymentioning

confidence: 99%

Little Albert’s alleged neurological impairment: Watson, Rayner, and historical revision.

Digdon

Powell

Harris

2014

History of Psychology

View full text Add to dashboard Cite

In 2012, Fridlund, Beck, Goldie, and Irons (2012) announced that "Little Albert"-the infant that Watson and Rayner used in their 1920 study of conditioned fear (Watson & Rayner, 1920)-was not the healthy child the researchers described him to be, but was neurologically impaired almost from birth. Fridlund et al. also alleged that Watson had committed serious ethical breaches in regard to this research. Our article reexamines the evidentiary bases for these claims and arrives at an alternative interpretation of Albert as a normal infant. In order to set the stage for our interpretation, we first briefly describe the historical context for the Albert study, as well as how the study has been construed and revised since 1920. We then discuss the evidentiary issues in some detail, focusing on Fridlund et al.'s analysis of the film footage of Albert, and on the context within which Watson and Rayner conducted their study. In closing, we return to historical matters to speculate about why historiographical disputes matter and what the story of neurologically impaired Albert might be telling us about the discipline of psychology today.

show abstract

Ethical Issues and Guidelines for Conducting Data Analysis in Psychological Research

Cited by 14 publications

References 28 publications

Detecting Duplication in Students’ Research Data: A Method and Illustration

Detecting Duplication in Students’ Research Data: A Method and Illustration

I Never Promised You a Rose Garden: When Research Questions Ought to Change

Little Albert’s alleged neurological impairment: Watson, Rayner, and historical revision.

Contact Info

Product

Resources

About