Papercheck is designed modularly, so you can add modules to check for anything. It comes with a set of pre-defined modules, and we hope people will share more modules.
Module List
You can see the list of built-in modules with the function below.
- all-p-values: List all p-values in the text, returning the matched text (e.g., ‘p = 0.04’) and document location in a table.
- all-urls: List all the URLs in the main text
- imprecise-p: List any p-values reported with insufficient precision (e.g., p < .05 or p = n.s.)
- llm-summarise: Generate a 1-sentence summary for each section
- marginal: List all sentences that describe an effect as ‘marginally significant’.
- osf-check: List all OSF links and whether they are open, closed, or do not exist.
- ref-consistency: Check if all references are cited and all citations are referenced
- retractionwatch: Flag any cited papers in the RetractionWatch database
- statcheck: Check consistency of p-values and test statistics
Built-in Modules
Below, we will demonstrate the use of each built-in module, first on
a single paper and then a list of papers, the psychsci list
of 250 open-access papers from Psychological Science.
paper <- psychsci$`0956797620955209`all-p-values
List all p-values in the text, returning the matched text (e.g., ‘p = 0.04’) and document location in a table.
all_p <- module_run(paper, "all-p-values")
all_p # print table| text | section | header | div | p | s |
|---|---|---|---|---|---|
| p = .003 | method | Questionnaires | 6 | 2 | 3 |
| p = .08 | method | Questionnaires | 6 | 2 | 3 |
| p < .001 | results | Results | 9 | 2 | 1 |
| p < .025 | results | Results | 9 | 2 | 3 |
| p = .040 | results | Results | 9 | 2 | 3 |
| p = .173 | results | Results | 9 | 2 | 3 |
| p = .006 | results | Results | 9 | 2 | 4 |
| p = .02 | results | Results | 9 | 2 | 4 |
| p = .691 | results | Results | 9 | 2 | 5 |
| p = .303 | results | Results | 9 | 2 | 5 |
| p = .023 | results | Results | 9 | 3 | 3 |
| p < .001 | results | Results | 9 | 3 | 3 |
| p = .006 | results | Results | 9 | 4 | 2 |
| p = .037 | results | Results | 9 | 4 | 2 |
| p = .038 | results | Results | 9 | 4 | 2 |
| p = .358 | results | Results | 9 | 4 | 2 |
| p < .001 | results | Results | 9 | 4 | 3 |
| p = .127 | results | Results | 9 | 4 | 3 |
| p = .062 | results | Results | 9 | 4 | 3 |
| p = .047 | results | Results | 9 | 4 | 3 |
Showing 20 of 20 rows
If you run this module on all 250 papers, you will get more rows than you probably want to print, so the default just shows you the first 20 rows.
all_p_ps <- module_run(psychsci, "all-p-values")
all_p_ps # print table| text | section | header | div | p | s | id |
|---|---|---|---|---|---|---|
| p < .001 | results | Results | 8 | 1 | 1 | 0956797613520608 |
| p < .001 | results | Results | 8 | 1 | 2 | 0956797613520608 |
| p < .001 | results | Results | 8 | 1 | 3 | 0956797613520608 |
| p = .016 | results | Results | 8 | 1 | 4 | 0956797613520608 |
| p < .001 | results | Results | 8 | 2 | 1 | 0956797613520608 |
| p < .001 | results | Results | 8 | 2 | 1 | 0956797613520608 |
| p = .005 | results | Results | 7 | 2 | 1 | 0956797614522816 |
| p < .001 | results | Results | 7 | 2 | 2 | 0956797614522816 |
| p = .153 | results | Results | 7 | 2 | 2 | 0956797614522816 |
| p < .001 | results | Results | 7 | 2 | 6 | 0956797614522816 |
| p < .001 | results | Results | 7 | 2 | 7 | 0956797614522816 |
| p = .315 | results | Results | 7 | 2 | 7 | 0956797614522816 |
| p = .024 | results | Eye movement strategies | 8 | 2 | 2 | 0956797614522816 |
| p = .008 | results | Eye movement strategies | 8 | 3 | 2 | 0956797614522816 |
| p = .09 | results | Eye movement strategies | 8 | 3 | 2 | 0956797614522816 |
| p = .028 | results | Eye movement strategies | 8 | 3 | 2 | 0956797614522816 |
| p = .122 | results | Eye movement strategies | 8 | 3 | 3 | 0956797614522816 |
| p = .001 | results | Eye movement strategies | 8 | 3 | 4 | 0956797614522816 |
| p = .237 | results | Eye movement strategies | 8 | 3 | 5 | 0956797614522816 |
| p = .008 | results | Gaze patterns | 9 | 2 | 1 | 0956797614522816 |
Showing 20 of 4683 rows
You can access the full table for further processing.
#> # A tibble: 6 × 2
#> text n
#> <chr> <int>
#> 1 p < .001 1511
#> 2 p < .05 152
#> 3 p < .01 143
#> 4 p = .001 123
#> 5 p = .002 99
#> 6 p < .0001 91all-urls
List all the URLs in the main text. There will, of course, be a few false positives when text in the paper is formatted as a valid URL.
all_urls <- module_run(paper, "all-urls")
all_urls # print table| text | section | header | div | p | s |
|---|---|---|---|---|---|
| 3.9.1.7 | method | Participants | 3 | 1 | 5 |
| https://osf.io/k2dbf | method | Analyses | 8 | 1 | 1 |
| https://osf.io/k2dbf | funding | Open Practices | 14 | 1 | 1 |
| https://osf.io/k2dbf | funding | Open Practices | 14 | 2 | 1 |
| http://www.psychologicalscience.org/publications/badges | funding | Open Practices | 14 | 2 | 3 |
Showing 5 of 5 rows
all_urls_ps <- module_run(psychsci, "all-urls")
# show the most common URLs
all_urls_ps$table |>
count(text, sort = TRUE) |>
head()#> # A tibble: 6 × 2
#> text n
#> <chr> <int>
#> 1 http://www.psychologicalscience.org/publications/badges 126
#> 2 http://pss.sagepub.com/content/by/supplemental-data 28
#> 3 https://osf.io/sh8qg 11
#> 4 https://osf.io/2j6ne 10
#> 5 https://osf.io/79kbm 10
#> 6 http://pss.sagepub.com/content/25/1/3.full 9imprecise-p
List any p-values that may have been reported with insufficient precision (e.g., p < .05 or p = n.s.).
imprecise <- module_run(paper, "imprecise-p")
imprecise # print tableYou may have reported some imprecise p-values
| text | section | header | div | p | s |
|---|---|---|---|---|---|
| p < .025 | results | Results | 9 | 2 | 3 |
Showing 1 of 1 rows
You can use the expand_text() function to get the full
sentence or paragraph for context. Here you can see that “p < .025”
was not an imprecisely reported p-value, but a description of the
preregistered alpha threshold.
exp <- expand_text(imprecise, paper)
exp$expanded # print expanded text#> [1] "The main effect of illness recency did not meet our preregistered threshold (p < .025)-recently ill: M = 661 ms, SD = 197; not recently ill: M = 626 ms, SD = 153, F(1, 400) = 4.23, η p 2 = .010, 90% CI = [.000, .039], p = .040-nor did the interaction between illness recency and face type (disfigured vs. typical), F(1, 400) = 1.87, η p 2 = .005, 90% CI = [.000, .027], p = .173."We can investigate the most common imprecise p-values in the PsychSci set. “p < .01” and “p < .05” are probably often describing figures or tables, but what is the deal with “p > .25”?
imprecise_ps <- module_run(psychsci, "imprecise-p")
imprecise_ps$table |>
count(text, sort = TRUE) |>
head()
[38;5;246m# A tibble: 6 × 2
[39m text n
[3m
[38;5;246m
We can expand the text to check the context for “p > .25”.
gt.25 <- imprecise_ps$table |>
filter(grepl("\\.25", text)) |>
expand_text(paper = psychsci)
gt.25$expanded[1:3] # look at the first 3#> [1] "Time increased negative attitudes among liberal participants, b = 0.13, SE = 0.06, 95% CI = [0.02, 0.24], t(1922) = 2.34, p = .020, but did not affect attitudes among conservative participants, b = -0.04, SE = 0.06, 95% CI = [-0.15, 0.07], t(1922) = -0.64, p > .250."
#> [2] "Time increased negative attitudes among liberal participants, b = 0.22, SE = 0.06, 95% CI = [0.10, 0.33], t(1922) = 3.63, p < .001, but did not affect attitudes among conservative participants, b = 0.03, SE = 0.06, 95% CI = [-0.09, 0.14], t(1922) = 0.45, p > .250."
#> [3] "The twoway interactions between knowledge distance and search effort (odds ratio = 1.67, p = .080) and between knowledge distance and search variation (odds ratio = 0.65, p = .077) were marginally significant, whereas the interaction between search effort and variation was not significant (odds ratio = 1.01, p > .250)."llm-summarise
Generate a 1-sentence summary for each section. This uses the
llm() function with the query “summarise this section
briefly, in one sentence” and returns a one-sentence summary of each
major section (abstract, intro, method, results, discussion,
acknowledgements, appendices)
summaries <- module_run(paper, "llm-summarise", seed = 8675309)
summaries # print default table| text | section | answer | time | tokens |
|---|---|---|---|---|
| Across taxa, animals possess a number of behavioral adaptations that function to mitigate the cos… | abstract | Humans have evolved behavioral adaptations to detect and avoid infectious microorganisms, includi… | 0.1327031 | 373 |
| the evolutionary-biology literature, which highlights the fact that organisms neutralize pathogen… | intro | The study aimed to replicate a previous finding that people who have recently been ill are more a… | 0.2404155 | 1406 |
| Method Except where noted, all methodological details-including all stimuli and dot-probe procedu… | method | The study replicated the methodological details of Miller and Maner (2011) to test the interactio… | 0.3714075 | 2314 |
| Results On the basis of our preregistered exclusion criteria, we removed the 9 participants with … | results | The study found a main effect of face type, with slower responses to disfigured faces, but no mai… | 0.2508097 | 1227 |
| Discussion The study replicated here has been interpreted as a key piece of evidence supporting a… | discussion | This study’s null findings do not provide strong evidence that pathogen avoidance does not vary a… | 0.2339141 | 904 |
| NA Funding This work was supported by Horizon 2020 European Research Council Grant No. StG-2015 6… | funding | This study was funded by a Horizon 2020 European Research Council Grant and has made its data, an… | 0.1644160 | 210 |
| Declaration of Conflicting Interests The author(s) declared that there were no conflicts of inter… | annex | The author(s) declared that there were no conflicts of interest with respect to the authorship or… | 0.0778079 | 101 |
Showing 7 of 7 rows
The default table isn’t very helpful here, so let’s customise the output. The object returned from a module usually contains a table. We just want to show the section and answer columns here.
summaries$table[, c("section", "answer")]#> section
#> 1 abstract
#> 2 intro
#> 3 method
#> 4 results
#> 5 discussion
#> 6 funding
#> 7 annex
#> answer
#> 1 Humans possess behavioral adaptations that help mitigate the costs of infectious microorganisms, including sensitivity to cues related to pathogens, which influence various psychological phenomena such as food learning, stigmatization, and mate choice.
#> 2 The study aimed to replicate a previous finding that people who have recently been ill are more attentive to visual cues of pathogens, and to explore whether this effect is related to self-reported disgust sensitivity and visual attention to pathogen cues.
#> 3 The study replicated the methodological details of Miller and Maner (2011) to test the interaction between illness recency and face type on attentional biases, using a dot-probe task with 413 participants and identical stimuli, procedures, and analyses as the original study.
#> 4 The study found a main effect of face type, with slower responses to disfigured faces, but no main effect of illness recency or interaction between illness recency and face type, and failed to replicate the original study's finding of attentional bias towards disfigured faces only in recently ill participants.
#> 5 This study's null findings do not provide strong evidence that pathogen avoidance does not vary as a function of immunological resistance due to methodological limitations, and the field would benefit from better developing the validity of measures used to test behavioral-immune system hypotheses.
#> 6 This study was funded by a Horizon 2020 European Research Council Grant and has made its data, analysis code, and design/analysis plans publicly available, earning Open Data and Preregistration badges.
#> 7 The author(s) declared that there were no conflicts of interest with respect to the authorship or the publication of this article.If you try to run this on the whole PsychSci set, you will get an
error message because the number of LLM calls is much higher than the
maximum number, which is set by default to 30 (the per-minute rate limit
for the default model on a free groq account). You can increase the
maximum allowable calls (llm() manages your rate limit, so
this would take nearly an hour at 30 calls per minute) using
llm_max_calls(), but we aren’t going to do this here.
summaries_ps <- module_run(psychsci, "llm-summarise", seed = 8675309)#> Error in (function (text, query, text_col = "text", model = llm_model(), : You need to include the argument API_KEY or set the variable GROQ_API_KEY in your Renvironmarginal
List all sentences that describe an effect as ‘marginally significant’.
marginal <- module_run(paper, "marginal")
marginal # print tableNo effects were described as marginally/borderline/close to significant.
Let’s check how many are in the full set.
marginal_ps <- module_run(psychsci, "marginal")
marginal_ps # print tableYou described effects as marginally/borderline/close to significant. It is better to write ‘did not reach the threshold alpha for significance’.
| text | section | header | div | p | s | id |
|---|---|---|---|---|---|---|
| Although the PTSD group showed a significant facilitation effect (M = 0.68), F(1, 15) = 5.575, p … | results | Details. | 11 | 1 | 4 | 0956797615569889 |
| A marginally significant negative correlation was found between suppression-induced forgetting on… | results | Posttraumatic Stress Diagnostic Scale | 13 | 1 | 4 | 0956797615569889 |
| When we more closely matched depression symptoms by comparing the low-BDI-II PTSD group with the … | results | The effects of depressive symptoms | 14 | 3 | 1 | 0956797615569889 |
| In that analysis, the group difference in suppression-induced forgetting only approached signific… | results | Conditionalized final-test data | 15 | 1 | 2 | 0956797615569889 |
| The Congruency × Alignment interaction was marginally significant for faces, F(1, 19) = 4.12, p =… | results | Results | 7 | 2 | 3 | 0956797615617779 |
| The twoway interactions between knowledge distance and search effort (odds ratio = 1.67, p = .080… | results | Results | 6 | 2 | 3 | 0956797616634665 |
| An omnibus 3 (sex ratio) × 2 (participant sex) ANOVA on vaccine-investment variance revealed a ma… | results | Results | 15 | 2 | 1 | 0956797616636631 |
| Further, we observed a marginally significant indirect effect of target condition on endorsement … | fig | 3 | NA | 2 | 0956797617692000 | |
| The dagger and asterisks indicate marginally significant and significant paths ( † p < .10, **p <… | fig | Fig. 5 . | 5 | NA | 7 | 0956797617692000 |
| Given the unexpected nature of the finding and the marginally significant trend, we will not inte… | annex | Supplemental Material | 20 | 1 | 2 | 0956797617716922 |
| A marginally significant fit effect was found for conscientiousness, but this Response surface in… | results | Actual Personality | 7 | 4 | 6 | 0956797617724435 |
| A fit bonus was not calculated for emotional stability because the a4 parameter did not reach the… | results | Actual Personality | 7 | 4 | 12 | 0956797617724435 |
| The fit bonus is reported only for those traits that had at least a marginally significant a4 par… | discussion | Discussion | 8 | 3 | 8 | 0956797617724435 |
| Participant height (added as a fixed effect to the interceptonly model) had a marginally signific… | results | Results | 9 | 6 | 1 | 0956797618772822 |
| There was a marginally significant interaction between participant height and opponent height on … | results | Results | 9 | 7 | 2 | 0956797618772822 |
| SES was marginally significantly related to coins kept for self in the dictator game versus the u… | results | Results | 9 | 10 | 6 | 0956797618772822 |
| Subsequently, opponent height was added to the model, which was marginally significantly related … | results | Studies 1 and 2: Combined Data Analysis | 14 | 2 | 2 | 0956797618772822 |
| However, results revealed a marginally significant interaction between implicit partner evaluatio… | results | Implicit partner evaluations and nonverbal behaviors | 13 | 2 | 3 | 0956797618785899 |
| Although the main effect of implicit partner evaluations on nonverbal behavior remained significa… | results | Implicit partner evaluations and nonverbal behaviors | 13 | 3 | 3 | 0956797618785899 |
| Following up on the debate on the use of significance levels in psychology, we empirically examin… | abstract | Abstract | 0 | 2 | 1 | 0956797619830326 |
Showing 20 of 97 rows
You can see that the default shows only the first 20 rows and
truncates long lines. We can adjust this by explicitly calling the print
function and setting maxrows and
trunc_cell.
marginal_ps |> print(maxrows = 5, trunc_cell = Inf)You described effects as marginally/borderline/close to significant. It is better to write ‘did not reach the threshold alpha for significance’.
| text | section | header | div | p | s | id |
|---|---|---|---|---|---|---|
| Although the PTSD group showed a significant facilitation effect (M = 0.68), F(1, 15) = 5.575, p = .032, η p 2 = .271, and the control group showed only a marginally significant effect (M = 0.56), F(1, 15) = 4.368, p = .054, η p 2 = .226, the difference between the groups was not significant, F(1, 30) = 0.100, p = .754, η p 2 = .003 (Fig. 3c). | results | Details. | 11 | 1 | 4 | 0956797615569889 |
| A marginally significant negative correlation was found between suppression-induced forgetting on the details measure and PDS scores in the PTSD group (Kendall’s τ = -.33, p = .07; Fig. 4b). | results | Posttraumatic Stress Diagnostic Scale | 13 | 1 | 4 | 0956797615569889 |
| When we more closely matched depression symptoms by comparing the low-BDI-II PTSD group with the high-BDI-II control group, we observed a marginally significant group-by-condition interaction for the details measure, F(1, 16) = 4.032, p = .062, η p 2 = .201. | results | The effects of depressive symptoms | 14 | 3 | 1 | 0956797615569889 |
| In that analysis, the group difference in suppression-induced forgetting only approached significance, F(1, 30) = 3.869, p = .058, η p 2 = .114. | results | Conditionalized final-test data | 15 | 1 | 2 | 0956797615569889 |
| The Congruency × Alignment interaction was marginally significant for faces, F(1, 19) = 4.12, p = .057, η p 2 = .18, and significant for line patterns, F(1, 19) = 85.53, p < .001, η p 2 = .82. | results | Results | 7 | 2 | 3 | 0956797615617779 |
Showing 5 of 97 rows
osf-check
List all OSF links and whether they are open, closed, or do not exist. This requires an internet connection to check the status of the link.
osf_check <- module_run(paper, "osf-check")
osf_check # print tableAll OSF links are open
| text | section | header | div | p | s | status |
|---|---|---|---|---|---|---|
| https://osf.io/k2dbf | method | Analyses | 8 | 1 | 1 | open |
| https://osf.io/k2dbf | funding | Open Practices | 14 | 1 | 1 | open |
| https://osf.io/k2dbf | funding | Open Practices | 14 | 2 | 1 | open |
Showing 3 of 3 rows
This would take a while to run for the 341 unique OSF links found in the PsychSci set (usually 1-2 seconds per link), so we will skip it here (but can reveal we found 10 closed links).
osf_check_ps <- module_run(psychsci, "osf-check")ref-consistency
Check if all references are cited and all citations are referenced.
ref_consistency <- module_run(paper, "ref-consistency")
ref_consistency # print tableThis module relies on Grobid correctly parsing the references. There may be some false positives. There are references that are not cited or citations that are not referenced
| bib_id | ref | missing |
|---|---|---|
| b0 | (Illness recency. 95 .05 .03 .06 .08 5. Latencies (typical faces; ms) 633.90 163.50. 06 .03 .07 …. | citation |
| b17 | Jones BC, Hahn AC, Fisher CI, Wang H, Kandrik M, Tybur JM, Debruine LM (2018). “Hormonal correlat… | citation |
| b18 | Jones BC, Hahn AC, Fisher CI, Wang H, Kandrik M, Tybur JM, Debruine LM (2018). “Reply to Fleischm… | citation |
| b32 | Tybur JM, Inbar Y, Aarøe L, Barclay P, Barlow FK, De Barra M, Žeželj .., I (2016). “Parasite stre… | citation |
| b35 | Tybur JM, Lieberman D, Kurzban R, Descioli P (2013). “Disgust: Evolved function and structure.” _… | citation |
Showing 5 of 5 rows
It looks like there are some references with missing citations. The first one doesn’t look like a reference, and grobid often parses tables oddly. You’d need to look at the original PDF to see if the others are actually missing or false positives. Here, they are all false positives, based on grobid not being able to match the in-text citation to the reference list.
retractionwatch
Flag any cited papers in the RetractionWatch database. It takes a substantial time to make calls to crossref for each reference, so we provide a summary of the database in papercheck, which is updated regularly (last 2025-02-28).
rw <- module_run(paper, "retractionwatch")
rw # print tableYou cited no papers in the Retraction Watch database (as of 2025-02-28)
In the full PsychSci set, we find three papers in the RetractionWatch database cited, one with a retraction and two with corrections.
rw_ps <- module_run(psychsci, "retractionwatch")
print(rw_ps, trunc_cell = 30) # print tableYou cited some papers in the Retraction Watch database (as of 2025-02-28). These may be retracted, have corrections, or expressions of concern.
| bib_id | doi | ref | id | retractionwatch | text |
|---|---|---|---|---|---|
| b15 | 10.1177/0956797612470827 | Kok BE, Coffey KA, Cohn MA,… | 09567976221150616 | Correction | The vagus nerve controls re… |
| b26 | 10.1186/gb-2013-14-10-r115 | Horvath S (2013). “DNA meth… | 09567976231194221 | Correction | Epigenetic aging is one suc… |
| b34 | 10.1038/s41562-023-01749-9 | Protzko J, Krosnick J, Nels… | 09567976241249183 | Retraction | One may also ask whether ou… |
| b25 | 10.1186/gb-2013-14-10-r115 | Horvath S (2013). “DNA meth… | 09567976241260247 | Correction | Discrepancies between someo… |
| b25 | 10.1186/gb-2013-14-10-r115 | Horvath S (2013). “DNA meth… | 09567976241260247 | Correction | EAA from the Horvath PanTis… |
| b25 | 10.1186/gb-2013-14-10-r115 | Horvath S (2013). “DNA meth… | 09567976241260247 | Correction | The replication of results … |
Showing 6 of 6 rows
Check the context of the citations in the text column:
rw_ps$table$text#> [1] "The vagus nerve controls respiratory sinus arrhythmia (RSA), the covariation of heart rate and respiration, known to be an index of cardiac health and predictor of enhanced capacity for emotion regulation (Butler et al., 2006;Kok et al., 2013)."
#> [2] "Epigenetic aging is one such indicator of accelerated biological aging (Horvath, 2013)."
#> [3] "One may also ask whether our results can be generalized from our CloudResearch and student samples to broader populations in the United States or elsewhere (Hanel & Vione, 2016;Protzko et al., 2023)."
#> [4] "Discrepancies between someone's biological and chronological age translate into \"age acceleration,\" with older biological age relative to chronological age indicating greater acceleration (Horvath, 2013)."
#> [5] "EAA from the Horvath PanTissue clock (Horvath, 2013) was calculated using the methylclock package (Pelegí-Sisó et al., 2021)."
#> [6] "The replication of results across the PedBE and Horvath PanTissue clocks increases our confidence in study findings because these tools share no overlapping DNAm sites (Horvath, 2013;L."statcheck
Check consistency of p-values and test statistics using functions from statcheck.
statcheck <- module_run(paper, "statcheck")
statcheck # print tableWe detected possible errors in test statistics
| test_type | df2 | test_comp | test_value | p_comp | reported_p | computed_p | raw | error | decision_error | one_tailed_in_txt | apa_factor | text | section | header | div | p | s |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| t | 248.4 | = | 2.01 | = | 0.023 | 0.0455124 | t(248.4) = 2.01, p = .023 | TRUE | FALSE | FALSE | 1 | Yes-the 90% confidence intervals of the difference in attentional bias for participants who were … | results | Results | 9 | 3 | 3 |
Showing 1 of 1 rows
Here we see a false positive, where the paper reported the results of an equivalence test, which are meant to be one-tailed, but statcheck did not detect that this was one-tailed.
In the full PsychSci set, there are more than 27K sentences with numbers to check, so this takes about a minute to run.
statcheck_ps <- module_run(psychsci, "statcheck")There will be, of course, some false positives in the full set of 151 flagged values. Let’s look just at the flagged values where the computed p-value is about double the reported p-value, and this changes the significance decision (at an alpha of 0.05).
statcheck_ps$table |>
filter(decision_error,
round(computed_p/reported_p, 1) == 2.0) |>
select(reported_p, computed_p, raw) |>
mutate(computed_p = round(computed_p, 4))#> reported_p computed_p raw
#> 1 0.0290 0.0589 F(1, 361) = 3.59, p = .029
#> 2 0.0470 0.0947 t(24) = 1.74, p = .047
#> 3 0.0270 0.0547 t(24) = 2.02, p = .027
#> 4 0.0400 0.0797 t(24) = 1.83, p = .040
#> 5 0.0480 0.0962 t(240) = 1.67, p = .048
#> 6 0.0460 0.0915 t(32) = 1.74, p = .046
#> 7 0.0420 0.0846 t(21) = 1.81, p = .042
#> 8 0.0343 0.0686 t(10) = 2.04, p = .0343
#> 9 0.0330 0.0654 t(55) = 1.88, p = .033