> ## Documentation Index
> Fetch the complete documentation index at: https://meta.niceshare.site/llms.txt
> Use this file to discover all available pages before exploring further.
# Crab Effect
> Crab Effect is a pull on a peer who starts to rise. A 2020 health-worker survey links a comparison score to that pull, and the path is small.
**Category**: Effects
**Type**: A pull, by others in the same group, on a person who starts to rise above them
**Origin**: The open bucket is a fisherman's story, not a count of crabs. A dated public use is Albert Shanker, "The Crab Bucket Syndrome," 19 June 1994. The counted survey is Burcu Uzum and colleagues, *Frontiers in Psychology*, 14 October 2022
**Also known as**: Crab mentality, crabs in a bucket, crab barrel syndrome, crab-bucket effect
**Quick Answer** — The Crab Effect is a pull, by others in the same group, on a person who starts to rise above them. The bucket with no lid is a story. It is not a count of crabs. In one 2020 survey of 302 health-sector workers, a higher social-comparison score went with a slightly lower score on a five-item crab scale. The path is −0.14. The average answer on that scale was 2.11 on a 1-to-5 scale, below the midpoint.
## What is the Crab Effect?
The Crab Effect is a pull, by others in the same group, on a person who starts to rise above them.
> "if I cannot have it, neither can you"
That line is how Barış Çavuş and Ruhi Sarpkaya name the mentality in their 2021 scale paper. It is a sentence about a wish. It is not a count of people who acted on the wish. A prize with one winner is a scarce slot. A skill that many people can learn is not. [Zero-Sum Thinking](/strategies/zero-sum-thinking) is the belief that one person's gain is another's loss. The Crab Effect is the further move of tugging the person who is ahead.
The tug is not a low expectation laid on someone before the work starts. That is the [Golem Effect](/effects/golem-effect). It is also not an advantage that grows because the person already ahead receives more. That is the [Matthew Effect](/effects/matthew-effect). The pages next door measure those other patterns.
### The Crab Effect in 3 Depths
* **Beginner**: Someone in your group does well, and the next remark tugs them down. Write the remark before you name a character.
* **Practitioner**: Write whether the good thing was scarce. One promotion is not a grade that many people can earn. Then write what was counted: a feeling on a survey, or an outcome.
* **Advanced**: In the 2020 health-worker survey, the path from the comparison score to the five-item score is −0.14, and the path from a Type A score is 0.12. Together the predictors account for 22 percent. The other 78 percent is not a result on this page.
## Origin
The open bucket is an anecdote. Çavuş and Sarpkaya retell a man who sees crabs in a bucket with no lid and asks why they do not escape. The fisherman says that one crab would get out, and that the others grab any crab that tries. No crab was counted in that story. This page does not treat the story as animal behavior.
A dated public use of the name is Albert Shanker, "Where We Stand: The Crab Bucket Syndrome," 19 June 1994. The text read here is the reprint in *American Educator*, Spring/Summer 1997. Shanker says teachers at Frank W. Ballou Senior High School in Washington, D.C., used the phrase for students who had given in to gangs and drugs and who pulled back a student trying to leave that life. He is summarizing Ron Suskind, *Wall Street Journal*, 26 May 1994. That article was not re-read.
In Shanker's summary, officials kept an awards assembly secret because they feared sneers. The assistant principal says it sends a terrible message, "that doing well here means you better not show your face." One honoree had to be ordered to the stage while other students shouted "Nerd!" Shanker also reports, from Suskind, that 20 percent of the sophomores who registered last September were gone by Thanksgiving. That 20 percent is a leaving figure for one school. It is not a count of students who pulled someone down. "Last September" is the column's phrase. This page does not turn it into a year.
Çavuş and Sarpkaya attribute an earlier school use to Daniel L. Duke, "Drift, detachment, and the need for teacher leadership," in D. R. Walling, ed., *Teachers as Leaders* (Phi Delta Kappa, 1994), pages 255–273. Their paraphrase says Duke treated the pattern as a stumbling block to teacher leadership, and that some schools resist the efforts of their members. The chapter was not re-read. No sentence from Duke is on this page.
## Key Points
Four readings keep the story, the survey, and the school scale apart. A wish printed in a paper is not a career that was blocked.
The fisherman's bucket is the picture Çavuş and Sarpkaya use to open their paper. One crab escapes. Several crabs do not, because each pulls the one above. That is the story. It is not a published count of climbing crabs, and it is not a license to call a group a bucket.
Uzum and Ozdemir's five-item scale, as used in the 2022 paper, includes "That my colleagues are more successful than me makes me anxious." Answers sit on a 1-to-5 scale. In the 302 health workers, the mean is 2.11 and the standard deviation is 0.76. The table's alpha is 0.76. The measures section cites the 2020 scale paper's alpha as 0.74. That 2020 paper was not re-read. A mean below the midpoint is not a workplace full of sabotage.
A single promotion, a single prize, or a single seat can make one person's rise costly for the next person. A skill, a grade with no cap, or a public compliment need not. [In-Group Bias](/effects/in-group-bias) is a preference for one's own group. The Crab Effect, when the surveys measure it, is a reported unease that someone inside that group is ahead. The health survey does not show that a colleague's success cut the respondent's pay.
Çavuş and Sarpkaya's school scale asks teachers in Aydın whether colleagues are jealous, gossip, or hinder them, and whether the school supports cooperation. The first three loadings they print are 0.844, 0.815, and 0.797. The two factors explain 32.688 percent and 26.317 percent of the item variance, 59.004 percent together. No student grade and no resignation is attached to those loadings.
## Applications
Use the name only after you have written the scarce slot, the thing that was counted, and the thing that was not.
Before you call a remark the Crab Effect, write what was scarce. One named successor is a slot. Praise that many people can receive is not. [Zero-Sum Thinking](/strategies/zero-sum-thinking) covers the belief. This page's surveys do not show that the belief produced a lost job.
In the 302 health workers, the five-item mean is 2.11 on a 1-to-5 scale. Most of the mass is not at the top of the scale. A story about a national character that skips this mean has left the file.
"At my school, there are teachers who are jealous of me when I achieve success" is a perception item. A class average, a promotion, and a resignation are outcomes. Çavuş and Sarpkaya measured the first kind in Aydın high schools in 2019–2020. They did not report the second kind.
If you are the person who rose, write the remark or the blocked credit on one line and the later outcome on another. Shanker's column gives a secret awards assembly and a shout of "Nerd!" Those are reported acts at one school. The 20 percent sophomore leaving figure is not the rate of those acts.
## Case Study
Between 1 September and 30 December 2020, Burcu Uzum, Yasemin Ozdemir, Semra Kose, Osman Seray Ozkan, and Okan Seneldir surveyed people working in the health sector in Turkey. The design is a convenience sample and an online questionnaire. A group of 50 had already checked that the questions read clearly. Those 50 are not inside the 302. Everyone who started the main form finished it, so the analyzed file is 302. The paper says 150 were men (49.7 percent) and 152 were women (50.3 percent). Eighty-eight people (29.1 percent) were ages 27 to 33, and 72 (23.8 percent) were ages 34 to 40. The other age bands are not printed here. One hundred thirty-seven (45.4 percent) had a bachelor's degree, 171 (56.6 percent) were married, and 226 (74.8 percent) worked in the private sector.
The question was whether a social-comparison score, and a Type A versus Type B score, would line up with the five-item crab score. The comparison scale is an 18-item Turkish adaptation. A high total is read as higher self-esteem. Its mean in this file is 4.79 (standard deviation 0.79), and its alpha is 0.93. Split by the authors' cut, 162 people are in the high group, mean 5.25 (standard deviation 0.51), and 140 are in the low group, mean 4.25 (standard deviation 0.42). The personality items are seven pairs. The paper says the points are multiplied by three, with a total below 100 read as Type B and a total above 100 read as Type A. The means they print beside those labels are 5.16 (standard deviation 0.62) for 163 people called Type A, and 3.45 (standard deviation 0.49) for 139 people called Type B. This page does not turn 5.16 into a total out of 100.
The zero-order correlations, marked at the 0.05 level, two-tailed, are 0.106 between comparison and personality, −0.129 between comparison and the crab score, and 0.282 between personality and the crab score. In the path model, the comparison score's path to the crab score is −0.14 (p = 0.03). The authors read a higher comparison score as a lower crab score. The Type A path is 0.12 (p = 0.01). They say the predictors explain 22 percent of the crab score. A contrast for Type B is p = 0.054, and a contrast for the low self-esteem group is p = 0.065. Neither is marked significant. The conclusion's broader sentence, that Type B is also an antecedent, is not what those two tests show.
The first measurement model did not fit as they wanted: chi-square 2140.30 on 505 degrees of freedom, RMSEA 0.08, CFI 0.88, p = 0.04, SRMR 0.064. They dropped four items with low loadings, SC7 at 0.254, SC13 at 0.401, AB6 at 0.201, and AB7 at 0.198. The revised model is chi-square 603.84 on 290 degrees of freedom, RMSEA 0.06, CFI 0.91, p = 0.04, SRMR 0.056. They call that acceptable. The chi-square p they print is still 0.04. The comparison scale's average variance extracted is 0.44. They cite a rule that a value under 0.50 can stand when composite reliability is above 0.60. Their composite reliability for that scale is 0.92.
The lesson is a small association in one survey. People who scored themselves higher on the comparison items scored themselves slightly lower on anxiety about colleagues' success, and people scored toward Type A scored slightly higher. The file does not assign a personality, does not watch anyone block a colleague, and does not follow a career. A health sector in Turkey in late 2020 is not a school, a family, or a national culture.
## Boundaries and Failure Modes
The pattern has a name, a story, and two instruments. It does not yet have an experiment that assigns the pull and counts the later harm.
The 302 are a convenience sample of health-sector workers who agreed to an online form. The authors chose that sector because they thought the work would bring more stress and higher self-esteem. That is a reason, not a finding. The paths are small. Type B at p = 0.054 and the low self-esteem split at p = 0.065 were not significant. Do not say this file showed that low self-esteem raises the crab score, or that Type B lowers it. Do not move −0.14, 0.12, or 22 percent onto a classroom, a household, or a firm.
The school instrument is a different object. Çavuş and Sarpkaya collected it in the 2019–2020 academic year from high school teachers in Aydın province, in Turkey's Aegean region. Exploratory analysis began with 320 forms, dropped 15 that failed their normality screen, and used 305. A 44-item draft lost two items with item-total correlations below 0.30 whose upper-versus-lower 27 percent means were not significant, so the factor analysis started from 42 items and ended at 27. The first Kaiser-Meyer-Olkin value on that 42-item set is 0.952, and Bartlett's chi-square is 8808.184, p below 0.01. Confirmatory analysis used a separate set: 298 forms, 19 dropped, 279 analyzed. The fit they print is chi-square divided by degrees of freedom equal to 2.56 (820.38/320), RMSEA 0.075, NFI 0.96, and NNFI, CFI, and IFI each 0.97. Standardized RMR is 0.057. GFI is 0.82 and AGFI is 0.79, which they call acceptable rather than perfect. Fifteen items are the individual factor and twelve are the organizational factor. After the exploratory step, alpha is 0.953 overall, 0.951 for the individual items, and 0.927 for the organizational items. A scale can be stable in one province and still not tell you whether a student learned less.
A common misuse is to treat Ballou's reported conditions as the size of the effect. The shooting, the assault, and the 20 percent of sophomores gone by Thanksgiving are in Shanker's summary of Suskind. They describe one school in 1994. They are not a rate of pull-down, and they are not a finding about a people. A [Self-Fulfilling Prophecy](/effects/self-fulfilling-prophecy) would be a belief that then produces the outcome it names. This page does not show that loop.
## Common Misconceptions
Three retellings travel with the bucket. One treats the fisherman as a biologist. One treats a questionnaire as a blocked career. One treats a non-significant split as a cause.
The story in Çavuş and Sarpkaya is a fisherman talking to a walker. One crab would leave. The others pull it back. No species, no number of crabs, and no timed climb is in that story. The human pattern, if it is present, has to be counted in people. The bucket is only the picture.
The Aydın instrument asks what teachers perceive. Loadings of 0.844, 0.815, and 0.797 sit on items about jealousy, about dislike of a principal's praise, and about being compared. The two factors explain 59.004 percent of the item variance. Explaining items is not explaining a promotion, a grade, or a resignation. None of those outcomes is in the scale paper.
The significant comparison path is −0.14 on the continuous score, p = 0.03. The authors read a higher score as lower crab tendency. The split they call low self-esteem is p = 0.065, and the Type B contrast is p = 0.054. Those two are not significant in the paper. A discussion sentence that low self-esteem increases the syndrome goes past those tests. The zero-order correlation of the comparison score with the crab score is −0.129. It is not the same object as −0.14, and it is not a cause.
## Related Concepts
The Crab Effect sits beside a belief that gains are scarce, beside a preference for one's own group, beside an advantage that compounds, beside a low expectation, and beside a belief that brings about what it expects.
The belief that one person's gain is another's loss. A tug on a peer who is rising can start from that belief. The health survey does not show a lost payment.
A preference for members of one's own group. The bucket in these papers is usually that group. Preferring the group is not the same as scoring 2.11 on the five-item scale.
An advantage that grows because the person already ahead receives more. The Crab Effect is the opposite picture: a pull back toward the group. Neither result is the other.
A low expectation that is followed by lower performance. The surveys here ask about jealousy and anxiety after someone is ahead. They are not that expectation.
A belief that helps produce the outcome it names. Shanker's column describes taunts at one school. It does not measure that loop.
## One-Line Takeaway
Before you call a group a bucket, write whether the good thing was scarce, and whether you counted a feeling or an outcome.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.