Selective Exposure in the Age of Social Media: Endorsements Trump Partisan Source Affiliation When Selecting News Online
2014 · paper · Media studies
Two experiments test how social endorsements affect which online news people choose. Stronger endorsements raised the probability that people selected content, and their presence reduced partisan selective exposure to levels indistinguishable from chance.
Checked: PDF from web.stanford.edu, fetched 5 Oct 2026. Page numbers are those of that PDF.
9 findings
Substance "(M = 0.30, SD = 0.27) ... (M = 0.27, SD = 0.21), conditions t(234) = .92, p = .18, one-tailed, Cohen's d = .57": the printed d does not correspond to the printed means and SDs or to the printed t and df (both give about 0.12); a stray word "conditions" also precedes the t statistic.
- p. 10 Study 1 Results, second paragraph
(M = 0.27, SD = 0.21), conditions t(234) = .92, p = .18, one-tailed, Cohen’s d = .57
- p. 10 Results, Democrats' MSNBC rate between conditions
conditions t(234) = .92, p = .18,
- p. 10 same sentence
one-tailed, Cohen’s d = .57
- p. 10 Results, Democrats' MSNBC rate
conditions t(234) = .92
Only the authors can say what is right here.
Judge's note: Read on the p. 10 image. From the means and SDs: the equal-weight pooled SD is the square root of (0.27² + 0.21²)/2 = 0.242, so d = 0.03/0.242 = 0.12; no weighting of SDs lying between 0.21 and 0.27 gets near .57, and rounding of the means could raise the difference to at most 0.04 (d about 0.17). From the test: 2 × 0.92 / √234 = 0.12. The t and p do correspond to each other (one-tailed p for t = .92 on 234 df is about .18), so the d is the odd value out. For comparison, the other three d values in the passage are roughly reproduced from their means and SDs: 0.17/0.242 = 0.70 (printed .72), 0.13/0.246 = 0.53 (printed .55), 0.10/0.246 = 0.41 (printed .36). The paper does not state a different effect-size formula for this comparison. Only the authors can say what value was intended. The stray 'conditions' is a separate S4 copy-edit point.
Second check: another model agreed
Report a problem with this finding
Reference or label In-text citations "Jacobson, 2000, 2005" (PDF p. 1) and "(McFadden, 1974)" (PDF p. 9) have no entry in the reference list.
- p. 1 Introduction, first paragraph
Jacobson, 2000, 2005
- p. 9 Analysis
based on (McFadden, 1974)
- p. 20 Reference list (J entries)
Jamieson, K. H., & Cappella, J. N. (2010).
Judge's note: Read the reference list on the p. 20 image: it runs Iyengar & Hahn (2009), Iyengar, Hahn, Krosnick, & Walker (2008), Jamieson & Cappella (2010), Kahneman, with no Jacobson entry. A search of the whole extracted text finds 'Jacobson' only at the p. 1 citation and 'McFadden' only at the p. 9 citation and the Table 1 row label 'McFadden R2'. The reviewers' side remark that Chaiken (1987) is listed but not cited was not made part of this verdict.
Second check: another model agreed
Report a problem with this finding
Reference or label Three citation defects: (a) "(McFadden, 1974)" is cited on p.9 but has no entry in the reference list; (b) "Mutz, D. C. (2004). Cross-cutting social networks ... American Political Science Review, 96, 111-126. doi:10.1017/S0003055402004264" (p.21, cited as "(Mutz, 2004)" on p.4) gives a year that does not fit its own volume and DOI, which point to 2002; (c) p.6 cites "(Silvia Knobloch-Westerwick, Sharma, Hansen, & Alter, 2005)" with a given name in the citation.
- p. 21 References
Mutz, D. C. (2004). Cross-cutting social networks: Testing democratic theory in practice. American Politi- cal Science Review, 96, 111-126. doi:10.1017/S0003055402004264
- p. 4 De facto and Purposive Isolation
(Mutz, 2004)
- p. 21 References
Mutz, D. C. (2004). Cross-cutting social networks: Testing democratic theory in practice. American Politi-
- p. 6 Hypotheses section
(Silvia Knobloch-Westerwick, Sharma,
Judge's note: (a) On the p.21 image the list runs Levendusky, Metzger, Mutz with no McFadden entry; no McFadden anywhere in the list on pp.19-22. Objective omission. (b) Entry read on the p.21 image. Internal evidence: the same list gives American Political Science Review volume 95 for 2001 twice (Hetherington, 95(3), doi ...S0003055401003045; Mutz & Martin, 95(1)), so volume 96 would be 2002, and the DOI of the Mutz entry has '02' where the 2001 entry has '01'. The inference assumes one volume per year; the actual publication year is outside the files, so the year is best put to the authors. (c) Printed as quoted in the p.6 text; the list entry is 'Knobloch-Westerwick, S., ...'. Cosmetic. Severity is set by (a) and (b).
Second check: another model agreed
Report a problem with this finding
Reference or label Panel references to Figure 3 are swapped: "Figure 3 (left) clearly shows that partisans were more than twice as likely to select an article with a strong social endorsement from a dissonant source" and "Figure 3 (right) provides evidence that they are not" (the between-condition comparison); in the printed figure the left panel is the condition comparison and the right panel is the partisan agreement breakdown.
- p. 11 Study 1 Results, text below Figure 2
Figure 3 (left) clearly shows that partisans were more than twice as likely
- p. 11 Study 1 Results, last paragraph
Figure 3 (right) provides evidence that they are not
- p. 12 Figure 3
Endorsement only condition
- p. 11 Results, p. 11
Figure 3 (right) provides evidence that they are not.
Judge's note: Page 12 image: the left panel has two facets, the first legibly 'Endorsement only condition' and the second '[S]oci[a]l & [par]tis[a]n condition'; the right panel has three facets ('[D]is[ag]reement', '[N]eit[h]er', '[Ag]reement') under a title ending 'mon[g] [pa]rtis[a]ns'. Fonts are partly missing but the panel identities are clear. The text on page 11 assigns the dissonant/agreeing-source result to the left and the endorsement-only versus both-cues comparison to the right, the reverse.
Second check: another model agreed
Report a problem with this finding
Reference or label Table 2 is titled "Table 2. News Simulation Interface (Study 2)", word for word the caption of the screenshot "Figure 4. News simulation interface (Study 2)"; the table is the mixed logistic regression of story selection.
- p. 16 Table 2 title
Table 2. News Simulation Interface (Study 2)
- p. 15 Figure 4 caption
Figure 4. News simulation interface (Study 2)
Judge's note: Pages 15 and 16 images. The text on page 15 describes Table 2 as 'the results of a mixed logistic regression model with a random intercept for each participant'. Table 1's title describes its model ('Mixed Logit Model of Article Choice, by Condition'); Table 2's title describes the figure instead.
Second check: another model agreed
Report a problem with this finding
Reference or label The entry "Conlisk, J. (1996). Why bounded rationality? Journal of Economic Literature, 34(2), 669-700. doi:10.1126/science.151.3712.867-a" (PDF p.20) carries a DOI in the journal Science's format, not one for the Journal of Economic Literature article.
- p. 20 References
Journal of Economic Literature, 34(2), 669-700. doi:10.1126/ science.151.3712.867-a
- p. 20 Reference list
doi:10.1126/ science.151.3712.867-a
Judge's note: Read on the p.20 image; printed as quoted. Internal evidence: the same prefix 10.1126/science is used in this list for the two Science articles (Salganik et al., doi:10.1126/science.1121066; Tversky & Kahneman, doi:10.1126/science.185.4157.1124, where the suffix is volume.issue.page of that entry), so the suffix 151.3712.867-a encodes volume 151, issue 3712, page 867-a, which does not match volume 34(2), pages 669-700. I did not resolve the DOI; what it actually points to is outside the files, but the identifier does not belong to the entry as printed.
Second check: another model agreed
Report a problem with this finding
Wording "Republicans ... (M = 0.41, SD = 0.43) than Democrats (M = 0.30, SD = 0.40), t(270) = 2.43, p = .008 in the condition with both social cues and partisan labels": 270 degrees of freedom need at least 272 Republicans and Democrats, but that condition has 299 subjects in all (Table 1) and the partisan analysis in the same condition reports "N = 187".
- p. 12 Study 1 Results, paragraph beginning 'We now turn'
t(270) = 2.43, p = .008 in the condition with both social cues and partisan labels
- p. 11 Study 1 Results, analysis of partisan identifiers in the both-cues condition
χ2(1, N = 187) = 15.09
- p. 13 Table 1, Nsubjects, Both
299
- p. 11 Figure 3 discussion
weaker endorsements χ2(1, N = 187) = 15.09
Only the authors can say what is right here.
Judge's note: Pages 10-13 images. A two-group t test has at most n1 + n2 − 2 df (fewer under Welch), so 272 partisan respondents are needed. The whole sample has 326 Democrats and 147 Republicans (page 10) across three conditions of 300, 111 and 299 subjects; about 19% are independents or other, and even counting leaners 81% of 299 is 242. The partisan χ2 tests in this condition give N = 187. The text says individual mean rates are used 'to avoid artificially inflating the number of observations', so the unit should be the respondent. No statement in the text or notes explains the df. Which analysis set or unit the test used is a question for the authors. The reviewer's side remark on missing commas and spacing ('d =.72') is trivial and not part of this verdict.
Second check: another model agreed
Report a problem with this finding
Wording The printed φ values do not correspond to the printed χ2 and N on any single basis: "χ2(1, N = 187) = 15.09, p < .001, φ = .14" and "χ2(1, N = 187) = 1.73, p = .19, φ = .14" share one φ despite the same N and a ninefold difference in χ2; "χ2(1, N = 299) = 23.12, p < .001, φ = .11" and "χ2(1, N = 111) = 3.89, p = .04, φ = .19" fit different bases.
- p. 11 Study 1 Results, paragraph on Figure 3
χ2(1, N = 187) = 15.09, p < .001, φ = .14
- p. 11 Study 1 Results, paragraph on Figure 3
= 1.73, p = .19, φ = .14
- p. 11 Results, dissonant source
15.09, p < .001, φ = .14
- p. 12 Results, both condition
23.12, p < .001, φ = .11
Only the authors can say what is right here.
Judge's note: Read on the page 11 and 12 images. With φ = sqrt(χ2/N): using the printed N, the four values are .28, .10, .19 and .28 (only the N = 111 one matches its printed .19). Using four choices per respondent (N × 4, the observation counts in Table 1), they are .14, .05, .09 and .14 (only the first matches). No basis gives .14 for both N = 187 tests, and neither basis gives .11 for the N = 299 test. The paper does not define how φ was computed; no explanation found in the text or notes. Worded as a question to the authors on which N each φ uses; the second .14 may be carried over from the first.
Second check: another model agreed
Report a problem with this finding
Wording Study 2 sample: the text says "leaving 141 respondents for the analysis", but the paired t test is "t(139) = 1.957" (140 participants) and Table 2 lists "Groups 140" in all three models.
- p. 14 Study 2, Sample
leaving 141 respondents for the analysis
- p. 14 Study 2, Results
t(139) = 1.957
- p. 16 Table 2, row Groups
140
- p. 16 Table 2
Groups 140
Only the authors can say what is right here.
Judge's note: Pages 14 and 16 images. 153 − 12 = 141 as stated. A paired t test with 139 df has 140 pairs, and Table 2 (random intercept per participant) has 140 groups in every column. The text, notes and table carry no mention of a further exclusion, so one participant is unaccounted for. Small, but the authors should say which number is the analysis sample.
Second check: another model agreed
Report a problem with this finding
22 minor (typos and the like; most checked by the judge only)
Wording Study 1: "Participants (N = 739)" but Table 1's N_subjects are 300, 111 and 299 (710 in total). Study 2: "leaving 141 respondents for the analysis" but Table 2 has Groups = 140 and the paired test is t(139).
- p. 9 A National Sample From an Online Crowd-Sourcing Service
Participants (N = 739) were recruited from Mechanical Turk
- p. 13 Table 1, row Nsubjects
300
- p. 9 A National Sample From an Online Crowd-Sourcing Service
Participants (N = 739)
- p. 13 Table 1
Nsubjects 300 111 299
Only the authors can say what is right here.
Judge's note: Table 1 on the p. 13 image gives N_subjects 300, 111, 299 and N 1,200, 444, 1,196, which is four choices per subject. The total is 710, 29 fewer than 739. I searched for an explanation. Note 3 (p. 18) says 'we include only those who answered correctly' the validation questions, but it is attached to the sample description and does not say whether 739 is before or after that exclusion. The Results say nonpartisans are included as a baseline, so missing party identification does not account for the gap. For Study 2, the p. 14 crop shows 153 less 12 removed leaving 141, while Table 2 on the p. 16 image shows Groups 140 in all three models and the text has t(139), which is 140 participants. Neither one-participant nor 29-participant difference is explained. The authors would need to say which counts were analysed.
Second check: another model agreed
Report a problem with this finding
Wording Notation shifts in the Analysis section: k is a second alternative in "U_ij − U_ik = (α_j − α_k) + β(x_ij − x_ik) + (γ_j − γ_k)z_i" and then "the item set, k"; j is both the alternative index and the number of alternatives ("out of j possible alternatives") while the sum runs to J; β and z_i become β_i and z_ik in the displayed model.
- p. 9 Analysis, utility difference
β(xij − xik) + (γj − γk)zi
- p. 9 Analysis, sentence before the equation
the item set, k; and the alternative, l, they choose out of j possible alternatives
Judge's note: Read on the p. 9 image. The utility line uses U_ij = α_j + βx_ij + γ_j z_i with k as the comparison alternative; the next sentence redefines k as the item set and l as the alternative. 'participant i with j choices' and 'out of j possible alternatives' use j as a count, whereas the denominator sums over j = 1 to J. The displayed model has β_i x_ikl and γ_l z_ik without saying why β gains a subscript i or z a subscript k. The intended model can be reconstructed, so this is a notational inconsistency and not a wrong equation.
Second check: another model agreed
Report a problem with this finding
Wording "In the social endorsements condition, however, the partisan alignment coefficients on source lose significance": the condition the paper names 'social endorsement' (Table 1 column 'Endorsement') showed no source labels and has no party-by-source coefficients; the coefficients meant are in the 'Both' column.
- p. 14 top of page
In the social endorsements condition, however, the partisan alignment coefficients on source lose significance
- p. 13 Table 1, column Endorsement
Endorsement
- p. 13 Table 1
Fixed effects Partisan Endorsement Both
- p. 13 Table 1
Democrat × Fox
Judge's note: Page 13 image: the Endorsement column has only Social endorsement and Order among the fixed effects; party × source terms appear only under Partisan and Both. Page 7 defines the 'social endorsement' condition as showing the endorsement cue but not the source label. The sentence on page 14 therefore names the wrong condition as written, though the intended meaning (the condition with both cues) is recoverable from the following clause. The reviewer's second point, about 'though not significant, the signs of all coefficients are in the expected direction', is not a defect: it refers to the coefficients other than Republican × Fox.
Second check: another model agreed
Report a problem with this finding
Wording Note 9 (PDF p.19) says pseudo R2 measures offer little utility and "We instead provide AIC and BIC", but Table 1 (p.13), to which the note is attached, prints a "McFadden R2" row (0.019, 0.016, 0.016) as well as AIC and BIC.
- p. 19 Note 9
We instead provide AIC and BIC.
- p. 13 Table 1
McFadden R2
Judge's note: Read note 9 on the p.19 image and Table 1 on the p.13 image: the McFadden R2 row is there, above LRT, AIC and BIC. The note marker 9 is at the end of the Study 1 results paragraph on p.14. Table 2 (p.16) has no pseudo R2 row. A minor mismatch between the note's 'instead' and the table; nothing is numerically wrong.
Second check: another model agreed
Report a problem with this finding
Typo The in-text citations read "Sears & Freeman, 1967" (PDF pp. 2, 3, 4, 7) while the reference list has "Sears, D. O., & Freedman, J. L. (1967)".
- p. 3 How Social Media Shape the Media Environment
Sears & Freeman, 1967
- p. 7 paragraph after Hypothesis 2
Sears & Freeman, 1967, pp. 209-212
- p. 21 References
Sears, D. O., & Freedman, J. L. (1967)
- p. 2 Introduction
as advocated by Sears & Freeman, 1967
Judge's note: Seen on the images: 'Sears & Freeman, 1967' on p. 2 and p. 3; the reference-list crop shows 'Sears, D. O., & Freedman, J. L. (1967). Selective exposure to information: A critical review. Public Opinion Quarterly, 31(2), 194-213.' The text layer has 'Freeman' four times (pp. 2, 3, 4, 7) and 'Freedman' only in the list. The p. 7 citation gives pp. 209-212, inside the listed range 194-213, so it is the same work with the name spelled two ways. Rated S4 because the reference can still be found; it is a misspelled name, not a wrong citation.
Second check: none (judge only)
Report a problem with this finding
Typo Missing or surplus small words: "selecting digital media (in the case music," (p. 6); "an individual tuned in to particular news program" (p. 2); "directly in homepage of the originating website" (p. 4); "say relevance to the one's profession" (p. 5).
- p. 6 last paragraph
selecting digital media (in the case music,
- p. 2 The Growing Importance of Social News
an individual tuned in to particular news program
- p. 4 first paragraph
directly in homepage of the originating website
- p. 5 Social Endorsements as More Powerful Heuristic Cues
relevance to the one’s profession
Judge's note: All four are as quoted on the images: p. 6 lacks 'of', p. 2 lacks 'a', p. 4 lacks 'the', and p. 5 has a surplus 'the'. Also seen on the p. 5 image: 'then discuss importance of the distinction'; and on p. 10: 'at higher rate than Republicans'. The p. 18 example a reviewer listed was not checked.
Second check: none (judge only)
Report a problem with this finding
Typo Reference list (PDF p.22): "OReilly, T. (2007). What is Web 2.0 ... Communications & strategies,17 (1)." lacks the apostrophe used in the citation "O'Reilly, 2007" (p.2), sits between Tversky & Kahneman and Valentino et al. instead of under O, and has a missing space after the comma.
- p. 22 References, between Tversky & Kahneman and Valentino et al.
OReilly, T. (2007). What is Web 2.0
- p. 2 The Growing Importance of Social News
O’Reilly, 2007
- p. 22 References
Communications & strategies,17 (1).
- p. 22 Reference list
OReilly, T. (2007).
Judge's note: Read on the p.22 image: the entry is printed 'OReilly' between Tversky and Valentino, with 'strategies,17 (1)'. The p.2 text cites "(O'Reilly, 2007)". Also true, as one note says, that Althaus & Tewksbury sits between the two Abramowitz entries on p.19 (seen on that page image).
Second check: none (judge only)
Report a problem with this finding
Typo "A 2011 study analyzing Neilsen data found that Facebook was the fastest-growing source of referrals": 'Neilsen' appears to be a misspelling of 'Nielsen'.
- p. 3 The Success of Socializing the News, second paragraph
A 2011 study analyzing Neilsen data
- p. 3
A 2011 study analyzing Neilsen data found that Facebook
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the audience-measurement company's name is spelled Nielsen. The p. 3 image shows 'Neilsen' as quoted. The paper does not spell the name anywhere else, so there is nothing internal to compare with.
Second check: none (judge only)
Report a problem with this finding
Typo "estimating the model via Newton-Ralphson maximum likelihood optimization": 'Ralphson' appears to be a misspelling of 'Raphson'.
- p. 10 Analysis, top of page
Newton-Ralphson maximum likelihood optimization
- p. 19 Note 9
We instead provide AIC and BIC.
- p. 3 Success of Socializing the News
analyzing Neilsen data
- p. 10 Analysis
Newton-Ralphson maximum likelihood
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the optimisation method is named Newton-Raphson. The p. 10 image shows 'Newton-Ralphson' as quoted. Two other points were bundled in by reviewers. 'Neilsen' (p. 3) is C003. Separately, and checkable inside the paper: note 9 (p. 19 crop) says pseudo R² measures 'offer very little utility for mixed logit models ... We instead provide AIC and BIC', yet Table 1 (p. 13 image) prints a 'McFadden R²' row (0.019, 0.016, 0.016) alongside AIC and BIC. That is a minor inconsistency between the note and the table, S3 at most, and is recorded here only as a remark.
Second check: none (judge only)
Report a problem with this finding
Typo Stray full stop inside a citation: "Iyengar, Hahn, Krosnick, & Walker., 2008".
- p. 5 para 1
Iyengar, Hahn, Krosnick, & Walker., 2008
- p. 5 Social media aggregators paragraph
Krosnick, & Walker., 2008
Judge's note: The p. 5 image shows '(Iyengar & Hahn, 2009; Iyengar, Hahn, Krosnick, & Walker., 2008)' with a full stop after 'Walker'.
Second check: none (judge only)
Report a problem with this finding
Typo In-text citation includes the first author's given name: "(Silvia Knobloch-Westerwick, Sharma, Hansen, & Alter, 2005)".
- p. 6 last paragraph
Silvia Knobloch-Westerwick, Sharma, Hansen, & Alter, 2005
- p. 5 first paragraph
Walker., 2008
- p. 6 paragraph before Hypothesis 2
(Silvia Knobloch-Westerwick, Sharma, Hansen, & Alter, 2005)
- p. 20 Reference list
Knobloch-Westerwick, S., Sharma, N., Hansen, D. L., & Alter, S. (2005)
Judge's note: The p. 6 image shows the citation with 'Silvia'; every other citation uses surnames only, and the reference list on p. 20 has 'Knobloch-Westerwick, S., Sharma, N., Hansen, D. L., & Alter, S. (2005)'. The stray full stop in 'Walker., 2008' that some reviewers added here is covered by C005.
Second check: none (judge only)
Report a problem with this finding
Typo Stray word 'conditions' before the test statistic: "(M = 0.27, SD = 0.21), conditions t(234) = .92, p = .18".
- p. 10 Study 1 Results, second paragraph
(M = 0.27, SD = 0.21), conditions t(234) = .92
- p. 10 Study 1 Results, second paragraph
conditions t(234) = .92
Judge's note: Read on the page 10 image. The sentence already says 'between the partisan label–only condition ... and the condition that also contains social endorsement cues'; the word 'conditions' after the closing parenthesis belongs to nothing.
Second check: none (judge only)
Report a problem with this finding
Typo "social endorsements proved to be a much stronger predictor of selection that did source cues": 'that' should be 'than'.
- p. 15 Conclusion, first paragraph
stronger predictor of selection that did source cues
- p. 15 Conclusion, first paragraph
a much stronger predictor of selection that did source cues
Judge's note: Read on the page 15 image, Conclusion, first paragraph.
Second check: none (judge only)
Report a problem with this finding
Typo Missing word: "the interfaces through they structure their customers' interaction with online content" should read 'through which they structure'.
- p. 17 Conclusion, second paragraph on the page
the interfaces through they structure their customers’
- p. 17 Conclusion, second paragraph on the page
the interfaces through they structure their customers’ interaction
Judge's note: Read on the page 17 image, second paragraph.
Second check: none (judge only)
Report a problem with this finding
Typo PDF p.17: "social and partisan cues presented that participants saw were comparable to how these cues exist on most websites" has two alternative wordings left in ("cues presented" / "cues that participants saw").
- p. 17 Conclusion, validity paragraph
social and partisan cues presented that participants saw were comparable
- p. 17 validity paragraph
social and partisan cues presented that participants saw
Judge's note: Read on the page 17 image; the sentence is printed exactly as quoted. Garbled wording, meaning still recoverable.
Second check: none (judge only)
Report a problem with this finding
Typo PDF p.17: "Though our sample comprised only of members of the population we care about" is ungrammatical ("was comprised only of" or "comprised only members").
- p. 17 validity paragraph
comprised only of members of the population we care about
Judge's note: Read on the page 17 image. As printed the clause has no well-formed verb phrase: 'comprised ... of' without 'was'. Cosmetic. The related 'comprising of both strong and weak ties' on p.4 is in the text but is a usage point only.
Second check: none (judge only)
Report a problem with this finding
Typo Note 6 (PDF p.18): "the tendency to chose one alternative over others" should read "to choose".
- p. 18 Note 6
the tendency to chose one alternative over others
Judge's note: Read on the page 18 image; 'chose' is printed. (One reviewer note mislabels the printed word as 'chance'; the defect is the same.)
Second check: none (judge only)
Report a problem with this finding
Typo PDF p.18: "with respect to what type of content shared, with what expectations and for what reasons" is missing a verb ("is shared").
- p. 18 last paragraph
what type of content shared, with what expectations
Judge's note: Read on the page 18 image; printed as quoted.
Second check: none (judge only)
Report a problem with this finding
Typo Note 11 (PDF p.19): "In order to maximize ecologically validity" should read "ecological validity".
- p. 19 Note 11
In order to maximize ecologically validity
- p. 19
In order to maximize ecologically validity, social endorsements
Judge's note: Read on the page 19 image; 'ecologically validity' is printed.
Second check: none (judge only)
Report a problem with this finding
Typo The reference list entry "Chaiken, S. (1987). The heuristic model of persuasion. Social Influence, p. 5." (PDF p.19) is not cited anywhere in the text or notes.
- p. 19 Reference list
Chaiken, S. (1987). The heuristic model of persuasion. Social Influence, p. 5.
Judge's note: Entry read on the p.19 image. Searched the full text for 'Chaiken': the only in-text occurrence is "(Axsom, Yates, & Chaiken, 1987)" on p.6, which is the separate Axsom entry. An uncited list entry; cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Reference list (PDF p.20): journal name printed "American psychologist, 58(9), 697" (Kahneman) against "American Psychologist, 39" (Kiesler et al.), and "CHI '03 Extended Abstracts on human Factors in Computing Systems" (Fogg) has a lower-case "human".
- p. 20 References
American psychologist, 58(9), 697
- p. 20 References
on human Factors in Computing Systems
Judge's note: Both read on the p.20 image; the Kiesler entry on the same page capitalises 'Psychologist'. Capitalisation slips.
Second check: none (judge only)
Report a problem with this finding
Typo Reference list (PDF p.21), Lee & Lee (1939): "A study of Father Coughlins Speeches" lacks the possessive apostrophe ("Coughlin's").
- p. 21 References, Lee & Lee
Father Coughlins Speeches
Judge's note: Read on the p.21 image; printed without an apostrophe. The possessive is evident from the phrase itself.
Second check: none (judge only)