Researchers misunderstand confidence intervals and standard error bars

Little is known about researchers' understanding of confidence intervals (CIs) and standard error (SE) bars. Authors of journal articles in psychology, behavioral neuroscience, and medicine were invited to visit a Web site where they adjusted a figure until they judged 2 means, with error bars, to be just statistically significantly different (p < .05). Results from 473 respondents suggest that many leading researchers have severe misconceptions about how error bars relate to statistical significance, do not adequately distinguish CIs and SE bars, and do not appreciate the importance of whether the 2 means are independent or come from a repeated measures design. Better guidelines for researchers and less ambiguous graphical conventions are needed before the advantages of CIs for research communication can be realized.     

Mea culpa: predicting stock prices from organizational attributions

People's causal attributions for events in their lives have been shown to relate to individual and interpersonal outcomes. Groups and organizations also make causal attributions, and this article examines whether their publicly communicated attributions predict organizational-level outcomes. By content analyzing attributions contained in corporate annual reports from 14 companies during a 21-year period, the authors found that organizations that made "self disserving" attributions- internal and controllable attributions for negative events-had higher stock prices 1 year later. The authors argue that claiming personal responsibility for negative events made the organizations appear more in control, leading to more positive impressions.     

Positivity can cue familiarity

Given that familiarity is closely associated with positivity, the authors sought evidence for the idea that positivity would increase perceived familiarity. In Experiment 1, smiling and thus positively perceived novel faces were significantly more likely to be incorrectly judged as familiar than novel faces with neutral expressions. In Experiment 2, subliminal association with positive affect (a positively valenced prime) led to false recognition of novel words as familiar. In Experiment 3, validity judgments, known to be influenced by familiarity, were more likely to occur if participants were in happy mood states than neutral mood states. Despite their different paradigms and approaches, the results of these three studies converge on the idea that, at least under certain circumstances, the experience of positivity itself can signal familiarity, perhaps because the experience of familiarity is typically positive.     

Research Report: Students' knowledge of abnormal psychology

The present study aims to compare whether final year psychology students (n = 26) could answer more items on a multiple choice questionnaire (MCQ) correctly on abnormal psychology than prospective psychology candidates (n = 77) and final year engineering students (n = 26). The three groups of students completed MCQs in five different fields of abnormal psychology namely; eating disorders, schizophrenia, anxiety disorders and depression, sexual disorders, and personality and compulsive disorders, respectively. They were also asked to indicate their confidence level in relation to the accuracy of the answers they had given. The results showed that final year psychology students scored higher on levels of accuracy, as well as confidence on all measures compared to the two other groups. On the subject of sexual disorders, engineering students scored higher than prospective psychology students. Implications of the present findings are discussed.     

Expression of Fos and Jun Proteins Following Passive Avoidance Training in the Day-Old Chick

It has been shown previously that the immediate-early genes, c-fos and c-jun mRNA are induced in the 1-day-old chick forebrain after one-trial passive avoidance training in which chicks learn to avoid pecking at a bitter-tasting bead. Here, we have studied the expression of their proteins using antibodies to Fos and Jun. Western blotting disclosed two immunoreactive bands for the anti-Fos antibody (47 and 54 kD) and two immunoreactive bands for the anti-Jun antibody (39 and 54 kD). Two hours post-training there was an increase in the number of Fos-positive stained nuclei in right intermediate medial hyperstriatum ventrale (IMHV) (P<0.01), left IMHV (P<0.05), right lobus parolfactorius (LPO) (P<0.025) and left LPO (P<0.05) of birds trained on the bitter bead compared with controls that had pecked a water-coated bead. Staining for Jun protein was significantly greater in the right LPO of trained chicks (P<0.01). Other forebrain regions showed no increase over quiet control levels. The findings are discussed in the context of the cascade of events involved in passive avoidance memory consolidation in the day-old chick.     

Being Polite and Keeping MUM: How Bad News is Communicated in Organizational Hierarchies1

There are many ways to communicate bad news. The MUM effect (Tesser & Rosen, 1975), which is keeping mum and not transmitting the bad news at all, is only one of many possible approaches. Using P. Brown and Levinson's (1987) politeness theory, an experimental study was conducted to address not whether bad news is being transmitted, but how bad news is being transmitted. The results show that most communicators tend to use politeness strategies when communicating bad news. Moreover, using politeness strategies to couch the message did not attenuate the informative value of the message. Focused contrasts revealed two strong interactions between gender and communication direction on strategy use. First, power differences more strongly predicted strategy use for men, whereas distance differences more strongly predicted strategy use for women. Second, men used most politeness strategies when the combined effect of power and distance was the highest (as the politeness theory would predict), but the trend was reversed for women. This finding suggests that politeness theory may not be an accurate model for describing female communicators.     

Recognizing familiar and unfamiliar faces

These experiments addressed why, in episodic-memory tests, familiar faces are recognized better than unfamiliar faces. Memory for faces of well-known public figures and unfamiliar persons was tested, not only with old/new recognition tests, in which initially viewed faces were discriminated from dis tractors, but also with tests of memory for specific information. These included: detail recall, in which a masked feature had to be described; orientation recognition, in which discrimination between originally seen faces and mirror-image reversals was required; and recognition and recall of labels for the public figures. Experiments 1 and 2 showed that memory for orientation and featural details was not robustly related either to facial familiarity or to old/new recognition rates. Experiment 3 showed that memory for labels was not the exclusive determinant of the famous-face advantage in recognition, since famous faces were highly recognizable even they were not labelable or when labels were forgotten. These results suggest that the familiarity effect, and face recognition in general, may reflect a nonverbal memory representation that is relatively abstract.     

Ingroup and outgroup inferences: when ingroup bias overwhelms outcome bias

Two experiments investigated conditions under which participants drew outcome-biased inferences about ingroups and outgroups. Participants read about ingroup and outgroup targets whose success or failure was influenced by an arbitrary decision rule. In Experiment 1, ingroup and outgroup members experienced two inconsistent outcomes (first success and then failure, or vice versa) despite almost identical performances. After reading about the first performance participants made outcome-biased inferences, but when the role of the decision rule became obvious inferences became group-serving. That is, outcomes continued to influence inferences when they cast the ingroup in a positive light (as when initial failure was followed by success) but failed to affect inferences when they were detrimental to the ingroup (as when initial success was qualified by later failure). In contrast, inferences about outgroups were outcome-biased when failure followed success, but not when success followed failure. The results of Experiment 2 showed that outcome biases influenced inferences when decision rules produced outcomes that promoted the ingroup but not when they produced outcomes that hurt the ingroup. No such benefit occurred for outgroups. The results confirm the impact of motivational concerns such as ingroup bias on the occurrence of outcome biases in inferences. © 1998 John Wiley & Sons, Ltd.     

An Attributional Analysis of Social Accounts: Implications of Playing the Blame Game1

These studies proposed that effective social accounts should contain external, unstable, specific, and uncontrollable attributions. In Study I, managers provided accounts for a negative event. The accounts contained highly unstable and specific attributions but, contrary to original predictions, they also contained highly internal and controllable attributions. In Study 2, managers rated the effectiveness of the account. Account effectiveness was positively related to unstable attributions, but was negatively related to external and uncontrollable attributions. These relationships were stronger (more negative) for accounts communicated downward than upward. These results indicate that blaming others or deflecting personal responsibility for negative events may be ineffective, especially when one is in a high‐status role.     

Investigating whether controlling and aggressive relationship behaviors are discriminant

Control is theorized as central to intimate partner aggression (IPA). Tools measuring nonphysical “controlling behaviors” in relationships have therefore been developed to identify the latent construct of control. However, the underlying assumption that “controlling behaviors” form a distinct subset of IPA has not been validated. This study investigates the divergent validity of acts considered as “controlling behaviors” against other aggressive acts used in relationships. The IPA and relationship literatures were reviewed to identify 1,397 items involving “controlling,” physical, sexual, and psychologically aggressive acts perpetrated and/or experienced by an intimate partner. In total, 101 item pairs were identified and used to measure IPA tactics across these categories. In Study 1, exploratory factor analysis in a community sample (N = 561) found no evidence of a distinct factor of “controlling behaviors.” Behaviors labeled as “controlling” in existing measures were distributed across other factors, including “eclectic aggression,” “direct psychological aggression,” and “monitoring acts.” In Study 2A (N = 424 students), confirmatory factor analysis replicated the results of Study 1 and established configural measurement invariance (Study 2B), indicating no evidence for psychometric differences between samples. These results indicate that behaviors described as “controlling” in existing measures were not statistically distinguishable from other forms of IPA, and suggest that future research should investigate motivational, rather than behavioral, differences in the use of IPA. The findings challenge research to confirm whether a set of discrete behaviors can be used to accurately identify control in relationships and question the validity of tools that adopt this methodology.