Psych 124 Flashcards
what is NOT recommended to decrease the rate of false-positive findings that researchers report:
researchers should only publish results that are statistically significant
They hypothesised that amateurs would score more highly than professionals. Their results found the opposite. They write up a paper for publication in which they state that they hypothesised that professionals would outscore amateurs. This is an example of which QRP
HARKing
In terms of null hypothesis significance testing (NHST), a false-positive result can be defined as:
The incorrect rejection of the null and the incorrect acceptance of the alternative hypothesis
The results obtained indicated that participants in the control group (read pop-culture magazines for 1 hour) were the least accepting of violence and displayed the most violent tendencies. This result was not what Alice was expecting, so she chose not to report this control condition in her paper. By not reporting this condition which QRP is Alice guilty of?
Outcome reporting bias
Various biases, cognitive and systemic, can lead to negative impacts in terms of the quality of research produced. From the following list, select all biases that would reduce the quality of research for a given domain:
a. Automation bias
b. Publication bias
c. Reporting bias
d. Gender bias
e. Citation bias
f. Confirmation bias
b,c,e,f
Publication bias
reporting bias
citation bias
confirmation bias
Conducting a rigorous meta-analysis can be extremely challenging. From the list below, select the features that can cause problems for a meta-analysis or otherwise reduce the quality of the meta-analysis:
a. Incomplete reporting
b. Publication bias
c. Outcome switching
d. Transparent materials and data
e. Studies published as registered report format articles
f. Following reporting guidelines, such as PRISMA.
a, b, c
incomplete reporting
publication bias
outcome switching
From the list below, select the examples that only apply to Registered Reports:
a. Methods can be submitted in full or in part.
b. Publication of the full study is guaranteed if the first version is accepted.
c. All the hypotheses must be in the first version.
d. Alterations from the first published version of the methods should be stated in the final published version.
e. It is still possible for p-hacking to take place.
b, c
publication of the full study is guaranteed
all hypotheses must be in the first version
According to Martinson et al. (2005) which of the following was not true?:
a. The most common questionable research practice was inadequate record keeping
b. Mid-career scientists were significantly more likely to report having engaged in some questionable research practices.
c. Failure to report an association with a funding body was more common than changing results to satisfy a funding body.
d. Falsifying research data was the least common questionable research practice
c
Failure to report an association with a funding body was more common than changing results to satisfy a funding body.
John et al. (2012) also looked at the prevalence of questionable research practices and found which of the following:
a. The prevalence of QRPs was not significantly different between domains of Psychology
b. Deciding to collect more or to collect less data than planned were both common.
c. Participants who were offered a sliding scale of charity donation admitted to more QRPs than those who were not
d. The most common QRPs were HARKing and p-hacking.
b,c
Deciding to collect more or to collect less data than planned were both common.,
Participants who were offered a sliding scale of charity donation admitted to more QRPs than those who were not
Anna is a developmental psychologist who conducted a study in 2015 on infants’ response to adult speech and found that 3-month-old infants prefer infant-directed speech to adult-directed speech. In 2020, she ran a study online using a modified experimental paradigm Carrying out a study in this way (with slight modifications) is an example of
A partial replication
Anna, in the study on infants’ preferences for different types of speech, despite the changes to the design, still found infants preferred for infant-directed speech. These results are most likely:
Generalisable
School A- improvement
School B – no change
School C – improvement
School D – worsening
School E – no change
Based on these results, the occupational therapist starts to write up their data and decides to publish only results from schools A, B and C. This is an example of
The file drawer problem
bias away from the null
Select any reasons why she might decide on a meta-analysis over a systematic review
a. A meta-analysis will include publishing the reasons for selection of studies to be included.
b. A meta-analysis might include unpublished studies as well as published ones.
c. In a systematic review, she might still be citing studies based on whether they agree with her predictions or not.
d. In a meta-analysis statistical comparisons will be carried out between datasets that may have different outcomes.
e. A meta-analysis must include more than 10 studies whereas a systematic review can include fewer than this.
d
In a meta-analysis statistical comparisons will be carried out between datasets that may have different outcomes
Various biases, cognitive and systemic, can lead to negative impacts in terms of the quality of research produced. From the following list, select all biases that would reduce the quality of research for a given domain:
a. Confirmation bias
b. Gender bias
c. Publication bias
d. Reporting bias
e. Citation bias
f. Automation bias
a, c, d, e
Confirmation bias
publication bias
reporting bias
citation bias
Which of the following types of files indicate good data sharing practices:
a. PDF data files
b. Meta-data files
c. Supplementary materials files hosted by the publishing journal
d. Raw data files
e. README files
f. Software-specific data files
b,d,e
raw data
meta data
README