А Rеsеаrсh Rероrt

  • Category:
  • Document type:
  • Level:
  • Page:
  • Words:


А Rеsеаrсh Rероrt

А Rеsеаrсh Rероrt Bаsеd On Dаtа Thаt Hаs Bееn Рrоvidеd


Hindsight bias, also known as the creeping determinism or the ‘knew-it-all-along effect, is often viewed as the inclination to view events that have taken place as being more predictable than, perhaps, they were before they occurred. Scholars see hindsight bias as a multifaceted phenomenon that easily affects processes, design, situations, and contexts. Many psychologists see hindsight bias as having the capacity to cause memory distortion in incidences where the recollection and subsequent reconstruction of content may lead to false or wring theoretical outcomes. Scholars have suggested that the hindsight bias effect has the ability to cause methodological problems when trying to understand, analyze or interpret experimental results in research studies. For instance, after an individual has been shown the results of an unforeseeable event, the individual believes that he ‘knew it all along’. Such instances are found in the writings of clinician recalling clinical trials or historians trying to describe the outcomes of war. It can, therefore, be posited that the events that may have been viewed as unpredictable in foresight may be judged as predictable when viewed in hindsight. Suffice to say, hindsight bias has the capacity to cloud judgments including medical diagnosis, legal decisions, sporting events, or election outcomes. Researchers have argued that trying to recall foresight knowledge often fails because the newly gained knowledge influences the memory by biasing the attempts to recall the foresight knowledge. This study seeks to investigate whether people with high self-presentation concerns show greater hindsight bias.

Literature Review

Knoll and Arkes, 2016). The subjects were then required to assess the probabilities of the possible outcome. It was found that the ‘after’ group exhibited higher probabilities to the events that they perceived as having taken place than the group that did not receive outcome information. The findings were seen to hold even when the research subjects were asked to answer as if they had not known the real outcome. (Knoll and Arkes (2016) indicate that the first experiment on hindsight bias was performed by a researcher called Fischhoff. The fischhoff experiment presented different subjects with various descriptions of several historical events. One group of research subjects were given four possible outcomes. They were required to estimate each probable outcome. Another group referred to as the after condition group were given the same descriptions, however, they were also given a fictitious possible outcome presented as an actual outcome of the events

The research design has variously been replicated, with different subject groups, for event description in many studies in various conditions; nevertheless, the findings have proved highly robust according to Pezzo (2011). Bias was found to exist for individual recall of prior probability. In the Fischhoff study, it was found that when some research subjects were requested to assign probabilities to the probable outcome of political events after the actual event had taken place, most subjects claimed that they had indicated higher probabilities than they had actually done.

Roese and Vohs (2012) opine that psychologists investigating the hindsight bias phenomenon have concluded that there are three levels that come into play in order to exhibit the overall effect including foreseability, inevitability, and memory distortion with each level influencing the other so as to arrive at the overall bias. The researchers give the example of a football game where some individuals were asked to estimate the probability of a win for their team. The estimates were 60% (Roese and Vohs, 2012). The team won. Later, when asked what they were likely to have said, they recall that they would have said closer to 80%. The differential between the two figures is what makes the hindsight bias. Essentially, an individual tricks himself to exhibit more confidence of the earlier unremembered estimation which amounts to the statement that “I said it would happen”. The failure to remember the prior prediction is the memory distortion (Roese and Vohs, 2012).

Inevitability occurs when individuals look back at events and easily identifies the cause and judge the outcome. For instance, a foreigner travelling in another country using a taxi, the taxi decides to take a long route to the individual’s destination. However, the foreigner doesn’t know the quickest route or how much it should cost. When later the individual gets to find out, he would make some links with a series of events and feel that he had been ripped off; however, the individual concludes that under the circumstances, such happenings were inevitable (Knoll and Arkes, 2016). The last level is the foreseability. Foreseability relates to an individual’s ability, knowledge, and beliefs. Often an individual is likely to hear of an event occurring and claim that he could have foreseen the event positing that ‘I could have told you that”

Renner (2003) classifies hindsight bias into the overconfidence and myopia. According to the author, myopia is an error in trying to understand the cause of an event by looking at the wrong place or exaggerating the effect of the right cause. Overconfidence, on the other hand, is brought about by overlooking other possible explanations and, consequently, leads to bad decisions for not learning from experience (Renner, 2003).

Research question

Having recognized that there are various levels of hindsight bias, the study seeks to find whether the bias is only found among the highly opinionated individuals or does it affect all people.

Question: do people with greater self-[presentation concerns show a greater hindsight bias?

Note: all three hypotheses proposed were supported.

  • Hypothesis 1: experimental group will demonstrate greater hindsight bias than control group.

  • Hypothesis 2: individuals with higher social desirability concerns show greater hindsight bias.

  • Hypothesis 3: individuals exposed to self-esteem threat display higher hindsight bias


The study enlisted 1391people, however, when the incomplete data was removed, 1160 responses remained. For inclusion in the study, the participants were expected to be fluent in English and be over 18 years.

Gender distribution


Valid Percent

Cumulative Percent

I prefer not to say

As indicated in table 1, there were more women than men in the study cohort. The individuals from the university were 236 which represented 20% of respondents; however 79% were drawn from outside the university. The age distribution was reported at between 18 and 78 years of age.


Valid Percent

Cumulative Percent




Other (please specify)

I prefer not to say

The ethnic spread of the participants was diverse with the Anglo-Australian leading with 602 participants or 51.9% while the Asian came second with 229 participants or 19.7% of the total study group. The participants from European origin made a sizeable group at 9.8 %. The other participants came from diverse groups including Africans, Arabs, and those who preferred not to state their ethnicity. The significance of the ethnic distribution is to demonstrate that hindsight bias cuts across the ethnic or racial divide and individuals across the globe are likely to behave the same in relation to hindsight bias.

Aim of the study

The study aim was to explore the hindsight bias, which in essence, is the propensity of individuals believing that they ‘knew it all along’. The hindsight bias was tested by requesting all the individual participant to state what they thought was the likelihood that Maria would make full recovery. Half of the participants were provided with information that was relevant to Maria’s diagnosis, prognosis, and distress level. The second group was given irrelevant information to Maria’s psychological health. Finally, all the respondents were required to recall the judgment they had made earlier regarding the likelihood of a full recovery. The respondents who changed their recollection of their original prognosis so that they could be aligned to the actual prognosis as provided were considered to show greater hindsight bias.

Ethical consideration

The ethical issues of the study had been approved by the university human research ethics committee and the participants were advised to present their ethical concern to the committee through the director for research ethics and integrity.


Results were obtained via an online survey platform (Qualtrics is a web-based survey platform which conducts surveys from creation, collection, and analysis) by subtracting the first recovery judgment from the second judgment. The survey questionnaire was emailed to the participants. Initially, the targeted participants were 1391 but only 1160 responded with completed questionnaire.


The study applied the Stober Social Desirability Scale (a score of 5 or lower = participant had low social desirability, a score of 11 or higher = participant had high social desirability). The respondents; after perusing Maria case study, to comment on whether Maria would make full recovery after attending psychological therapy. Half of the participants were provided with information that was relevant to Maria’s diagnosis, prognosis, and distress level. The second group was given irrelevant information to Maria’s psychological health. Finally, all the respondents were required to recall the judgment they had made earlier regarding the likelihood of a full recovery.

The dependent variable is size of the hindsight bias (the difference between the initial hindsight bias and recollection bias).The independent variable is Maria’s recovery, scores on the social desirability scale, and the threat to self-esteem.

Descriptive statistics

Social Desirability (SD)


Valid Percent

Cumulative Percent

The table show that the participants with social desirability represented 20.7% on the scale with a 44.1 valid percentage while those with high social desirability had a valid score 26.2%. 53.1% of the total participants did not reflect any valid percentage.

Independent Samples Test

Levene’s Test for Equality of Variances

t-test for Equality of Means

Sig. (2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

hindsight bias index

Equal variances assumed



Equal variances not assumed



The table shows the independent sample test which shows that the t=value is significant at 6.75 on assumed equal variances and 6.9 on the equal variances not assumed. However, the mean difference for both measures is insignificant at <1.1. The confidence level also reflects significant difference with both low and upper scales with the low reflecting 1.5 against the high at 0.8.


The study results show that the hindsight bias measured on the social desirability scale confirms the assumption that individuals are likely to follow the three levels of bias, the inevitability and foreseability. However, the results reveal that hindsight bias is not affected by culture, but is an innate human phenomenon even though social desirability and self-esteem vary across cultures, gender, ages, etc. The results also reveal that hindsight bias is not a generational phenomenon. Understanding hindsight bias will help reduce it, and this is important because hindsight is manifested the real world and across the global society.


Bernstein, D. & Atance, C. (2007). Hindsight Bias and Developing Theories of Mind Child Development, 78 (4), pp. 1374–1394.

, 25(1), pp. 1-9.Social CognitionBlank, H., Musch, J., & Pohl, R. F. (2007). Hindsight Bias: On Being Wise After the Event,

, 12, pp. 304-312. doi:10.1136/qhc.12.4.304.Quality and Safety in Health CareFischhoff, B. (2003). Hindsight and foresight: The effect of outcome knowledge on judgment under uncertainty,

Hoboken: John Wiley & Sons. Hardman, D & Hardman, D. (2009). Judgment and Decision Making: Psychological Perspectives,

Harley, E. (n. d.). The “Saw-it-All-Along” Effect: Demonstrations of Visual Hindsight Bias. Retrieved from:

doi: 10.1002/bdm.1950Journal of Behavioral Decision Making.Knoll, M. A., & Arkes, H. R. (2016). The effects of expertise on the hindsight bias,

23, pp. 131-160.Journal of Behavioral Decision Making, Marks, A. & Arkes, H. (2010). The effects of mental contamination on the hindsight bias: Source confusion determines success in disregarding knowledge,

11 (4-5), pp. 421–41.Memory, Pezzo, M. (2003). Surprise, Defense, or Making Sense: What Removes Hindsight Bias?

5 (9), pp. 665–78.Social and Personality Psychology Compass Pezzo, M. (2011). Hindsight Bias: A Primer for Motivational Researchers,

, 8 (1), pp. 72–98.Journal of Empirical Legal StudiesRachlinski, J., Guthrie, C. & Wistrich, A. (2011). Probable Cause, Probability, and Hindsight,

, 11, 455-472. doi: 10.1080/09658210244000531MemoryRenner, B. (2003). Hindsight bias after receiving self-relevant health risk information: A motivational perspective.

, 7, 411-426. doi: 10.1177/1745691612454303Perspectives on Psychological ScienceRoese, N. & Vohs, K. (2012). Hindsight bias,

, 17(3), pp. 222.European Journal of Psychological AssessmentStöber, J. (2001). The Social Desirability Scale-17 (SDS-17): Convergent validity, discriminant validity, and relationship with age,

Thaler, R. & Sunstein, C. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness, New Haven: Yale University Press.

Wilcox, C. (2011). Bias: the unconscious deceiver, Bloomington: Xlibris Corporation.

Zeynep, C. (2015). Handbook of research on behavioral finance and investment strategies: decision making in the financial industry: decision making in the financial industry, Hershey: IGI Global.