4e. Nomological network

NoteWhere am I in the pipeline?
  • Inputs: Preregistered hypothesis grid in validation_hypotheses_prereg.csv, wrangled S5 items and S5 scales from Study 5, cached S5 validation outputs (outputs/tables/s5_validation_results.csv, outputs/tables/s5_cor_matrix.csv), and S4 country-adjusted anchor correlations exported by Study 4 to outputs/tables/s4_external_anchor_correlations.csv.
  • This page: Nomological network (Stage 8). Synthesises external-validity evidence: the preregistered hypothesis verdict roll-up by DisInformeter construct, the S5 nomological-network correlogram against external anchor scales, the Fisher-z TOST equivalence summary for the <|.30| rows of the preregistration, and the S4 country-adjusted civic / political and misinformation anchor panel. This page is non-criterion: it carries no FIMI prediction component.
  • Hands off to: 4a Measurement architecture (construct revision when external evidence is weak) · 4c FIMI prediction (criterion-side evidence) · 4d Deployment and invariance (short-form transportability).

Overview

The five cross-study pages partition the eight-stage pipeline. This page (4e) owns Stage 8 — the joint nomological-network and external-validity synthesis. It consumes the Study 5 preregistered hypothesis grid (validation_hypotheses_prereg.csv), runs the same run_validation_tests() pass the per-study page does (no lavaan refit anywhere on this page), and rolls the resulting verdicts up to a construct-by-construct decision table. It also reads the S4 country-adjusted external-anchor export from the Study 4 page, giving the Stage 8 synthesis both a preregistered Lithuanian validation wave and a multinational anchor panel.

The construct, item, and operator vocabulary used here is identical to the per-study pages. For the =~ / <~ / ~ distinction, see the operator legend on 04a.

Preregistered verdict roll-up

The preregistration registers, for every DisInformeter construct × external anchor pair, a directional prediction (+ / −), a small-effect-bound equivalence prediction (<|.3|), or a u-shape prediction. run_validation_tests() in R/validation_helpers.R operationalises the four-verdict scorecard:

  • Confirmed — directional + / −: one-sided p < .05 in the predicted direction; u-shape: positive curvature with p < .05.
  • Disconfirmed — directional prediction with the observed correlation in the wrong direction (one-sided p > .50), or u-shape with quadratic coefficient not significantly positive.
  • Equivalence supported — <|.3| row where Fisher-z TOST rejects both ±.30 bounds (p < .05) and the 90 % CI lies inside (−.30, .30).
  • Equivalence inconclusive — <|.3| row where TOST fails to reject one or both bounds.
  • Inconclusive — directional row with a CI crossing zero (p_one ∈ [.05, .50]).

The table below rolls those verdicts up to one row per DisInformeter construct. Each cell is a count of primary preregistered hypotheses (item-subset and u-shape secondary echoes are excluded) — a construct with seven anchors and a single u-shape secondary contributes seven primary rows here. The right-most “share confirmed/equivalence-supported” column sorts the table so constructs with the strongest external profile sit at the top.

Table 1: Preregistered hypothesis verdict roll-up by DisInformeter construct (Study 5; primary rows only, n = 149 preregistered tests). Built from `outputs/tables/s5_validation_results.csv` via `run_validation_tests()`.
Construct Tests Confirmed Equivalence supported Equivalence inconclusive Inconclusive Disconfirmed % confirmed/equiv-sup
Exploitation 13 11 2 0 0 0 100%
Abandoned [general] 7 7 0 0 0 0 100%
Abandoned by media (censorship) 6 4 2 0 0 0 100%
Superiority [general] 5 5 0 0 0 0 100%
General 15 12 0 0 3 0 80%
Threat [general] 10 8 0 1 1 0 80%
Abandoned by own state 4 2 1 1 0 0 75%
Pragmatism China 15 9 2 1 2 1 73%
Superiority Russia 11 7 1 2 1 0 73%
Immigration 7 3 2 1 1 0 71%
Unprotected by international allies 9 5 1 1 2 0 67%
Pragmatism Russia 17 8 2 3 4 0 59%
Queer Sentiment 7 4 0 3 0 0 57%
Poisonous ethnocentrism 8 2 2 0 1 3 50%
Superiority China 8 2 2 2 2 0 50%
Fear [general] 7 1 2 0 2 2 43%
Table 2

Untested rows. None; every preregistered cell with non-NA prediction had sufficient variance and matching columns in study5_scales.parquet.

S5 nomological-network correlogram

The heatmap below mirrors the Study 5 page’s static export (the interactive plotly widget lives on the Study 5 page; this synthesis page keeps the print-ready static view so the prereg-direction violations are visible at a glance). Cells are Pearson r on the primary preregistered rows. Cells outlined in red contradict the preregistered prediction (+ with r ≤ 0, − with r ≥ 0, or <|.3| / omnibus with |r| ≥ .30).

Figure 1: Study 5 nomological-network correlations (primary preregistered rows). Built from s5_validation_results.csv; red outlines mark cells whose observed Pearson r contradicts the preregistered prediction.

TOST equivalence summary

The preregistration distinguishes between rows where a zero correlation is the substantive expectation — anti-elitism × queer sentiment, NATO feeling thermometer × generalised mistrust, and the other <|.3| cells — and rows where a directional effect is predicted. For the equivalence rows, “fails to reach significance” is not informative: the prereg has to demonstrate the absence of a meaningful association. The Fisher-z TOST tests, with both bounds at ±.30 against the observed Pearson r, are the relevant decision rule (.tost_correlation() in R/validation_helpers.R). The table below lists every primary <|.3| row, the observed correlation with its 90 % CI (matching the TOST decision rule), the larger of the two one-sided p-values (the TOST p), and the verdict.

Table 3: Fisher-z TOST equivalence verdicts for the preregistered `
Construct Anchor Family n r 90% CI TOST p Verdict
Abandoned by own state panas_na Omnibus discriminant 247 0.17 [0.05, 0.29] 0.411 Equivalence inconclusive
Immigration itt_lgbt Equivalence (`<&#124;.3&#124;`) 248 0.43 [0.32, 0.53] 0.990 Equivalence inconclusive
Pragmatism China feel_ru Equivalence (`<&#124;.3&#124;`) 247 0.37 [0.26, 0.47] 0.898 Equivalence inconclusive
Pragmatism Russia feel_ch Equivalence (`<&#124;.3&#124;`) 247 0.41 [0.30, 0.51] 0.978 Equivalence inconclusive
Pragmatism Russia itt_lgbt Equivalence (`<&#124;.3&#124;`) 248 0.58 [0.49, 0.66] 1.000 Equivalence inconclusive
Pragmatism Russia panas_pa Omnibus discriminant 248 0.15 [0.03, 0.27] 0.171 Equivalence inconclusive
Queer Sentiment itt_imi Equivalence (`<&#124;.3&#124;`) 248 0.46 [0.36, 0.56] 0.999 Equivalence inconclusive
Queer Sentiment panas_na Omnibus discriminant 247 −0.13 [−0.25, −0.01] 0.062 Equivalence inconclusive
Queer Sentiment panas_pa Omnibus discriminant 248 0.18 [0.06, 0.30] 0.597 Equivalence inconclusive
Superiority China feel_ru Equivalence (`<&#124;.3&#124;`) 247 0.39 [0.27, 0.49] 0.936 Equivalence inconclusive
Superiority China panas_pa Omnibus discriminant 248 0.14 [0.01, 0.26] 0.082 Equivalence inconclusive
Superiority Russia feel_ch Equivalence (`<&#124;.3&#124;`) 247 0.54 [0.44, 0.62] 1.000 Equivalence inconclusive
Superiority Russia panas_pa Omnibus discriminant 248 0.17 [0.05, 0.29] 0.375 Equivalence inconclusive
Threat [general] cmq Equivalence (`<&#124;.3&#124;`) 248 0.37 [0.25, 0.47] 0.874 Equivalence inconclusive
Unprotected by international allies panas_na Omnibus discriminant 247 0.17 [0.05, 0.29] 0.332 Equivalence inconclusive
Abandoned by media (censorship) panas_na Omnibus discriminant 247 0.01 [−0.11, 0.14] 0.000 Equivalence supported
Abandoned by media (censorship) panas_pa Omnibus discriminant 248 0.12 [0.00, 0.25] 0.043 Equivalence supported
Abandoned by own state panas_pa Omnibus discriminant 248 0.07 [−0.06, 0.19] 0.002 Equivalence supported
Exploitation panas_na Omnibus discriminant 247 0.09 [−0.03, 0.21] 0.007 Equivalence supported
Exploitation panas_pa Omnibus discriminant 248 0.11 [−0.02, 0.23] 0.019 Equivalence supported
Fear [general] itt_imi Equivalence (`<&#124;.3&#124;`) 248 0.09 [−0.03, 0.21] 0.000 Equivalence supported
Fear [general] itt_lgbt Equivalence (`<&#124;.3&#124;`) 248 0.12 [0.00, 0.24] 0.002 Equivalence supported
Immigration panas_na Omnibus discriminant 247 −0.04 [−0.17, 0.08] 0.000 Equivalence supported
Immigration panas_pa Omnibus discriminant 248 0.02 [−0.11, 0.14] 0.000 Equivalence supported
Poisonous ethnocentrism panas_na Omnibus discriminant 247 −0.06 [−0.19, 0.06] 0.001 Equivalence supported
Poisonous ethnocentrism panas_pa Omnibus discriminant 248 0.12 [0.00, 0.24] 0.037 Equivalence supported
Pragmatism China panas_na Omnibus discriminant 247 −0.05 [−0.17, 0.08] 0.001 Equivalence supported
Pragmatism China panas_pa Omnibus discriminant 248 0.11 [−0.02, 0.23] 0.019 Equivalence supported
Pragmatism Russia itt_imi Equivalence (`<&#124;.3&#124;`) 248 0.18 [0.05, 0.30] 0.021 Equivalence supported
Pragmatism Russia panas_na Omnibus discriminant 247 −0.07 [−0.19, 0.06] 0.002 Equivalence supported
Superiority China itt_imi Equivalence (`<&#124;.3&#124;`) 248 0.12 [−0.01, 0.24] 0.001 Equivalence supported
Superiority China panas_na Omnibus discriminant 247 0.00 [−0.12, 0.13] 0.000 Equivalence supported
Superiority Russia panas_na Omnibus discriminant 247 −0.03 [−0.15, 0.10] 0.000 Equivalence supported
Unprotected by international allies panas_pa Omnibus discriminant 248 −0.06 [−0.18, 0.07] 0.001 Equivalence supported

S4 political, civic, and misinformation anchors

The S4 wave administers civic / political anchors and misinformation-perception anchors alongside the deployed multinational DisInformeter short form. The Study 4 page exports country-adjusted Pearson correlations for these anchors to outputs/tables/s4_external_anchor_correlations.csv; those estimates are read here so the Stage 8 synthesis can compare S5’s preregistered Lithuanian validation network with S4’s multinational anchor evidence.

Table 4
Largest absolute S4 country-adjusted external-anchor correlations by family and DisInformeter predictor. Full export: outputs/tables/s4_external_anchor_correlations.csv.
Family Predictor Anchor n country-adjusted r FDR p pooled r
Civic / political ABAND Institutional trust 8040 −0.37 <0.001 −0.41
Civic / political ABAND External political efficacy 8040 −0.25 <0.001 −0.28
Civic / political ABAND Life satisfaction 8040 −0.22 <0.001 −0.25
Civic / political FEAR Democracy importance 8040 0.16 <0.001 0.15
Civic / political FEAR Religiosity 8040 0.10 <0.001 0.11
Civic / political FEAR External political efficacy 8040 −0.07 <0.001 −0.07
Civic / political GEN Institutional trust 8040 −0.17 <0.001 −0.23
Civic / political GEN Religiosity 8040 0.16 <0.001 0.22
Civic / political GEN EU support 8040 −0.15 <0.001 −0.15
Civic / political SUPF Democracy importance 8040 −0.24 <0.001 −0.25
Civic / political SUPF Religiosity 8040 0.23 <0.001 0.28
Civic / political SUPF Left-right self-placement 8040 0.16 <0.001 0.12
Civic / political SUPFCH Religiosity 8040 0.18 <0.001 0.23
Civic / political SUPFCH Internal political efficacy 8040 0.18 <0.001 0.18
Civic / political SUPFCH Generalised social trust 8040 0.18 <0.001 0.14
Civic / political THREAT EU support 8040 −0.26 <0.001 −0.25
Civic / political THREAT Left-right self-placement 8040 0.25 <0.001 0.23
Civic / political THREAT Institutional trust 8040 −0.24 <0.001 −0.27
Misinformation ABAND Perceived misinfo creators 8040 0.33 <0.001 0.33
Misinformation ABAND Foreign-actor topic salience 8040 0.33 <0.001 0.33
Misinformation ABAND Perceived misinfo impact 8040 0.18 <0.001 0.18
Misinformation FEAR Responsibility for prevention 8040 0.28 <0.001 0.28
Misinformation FEAR Perceived misinfo impact 8040 0.26 <0.001 0.25
Misinformation FEAR Perceived misinfo creators 8040 0.21 <0.001 0.21
Misinformation GEN Foreign-actor topic salience 8040 0.31 <0.001 0.31
Misinformation GEN Perceived misinfo creators 8040 0.30 <0.001 0.30
Misinformation GEN Perceived misinfo impact 8040 0.22 <0.001 0.20
Misinformation SUPF Foreign-actor topic salience 8040 0.20 <0.001 0.19
Misinformation SUPF Responsibility for prevention 8040 −0.11 <0.001 −0.11
Misinformation SUPF Perceived misinfo creators 8040 0.10 <0.001 0.09
Misinformation SUPFCH Foreign-actor topic salience 8040 0.16 <0.001 0.16
Misinformation SUPFCH Perceived misinfo creators 8040 0.09 <0.001 0.09
Misinformation SUPFCH Perceived misinfo impact 8040 0.09 <0.001 0.08
Misinformation THREAT Perceived misinfo creators 8040 0.31 <0.001 0.32
Misinformation THREAT Foreign-actor topic salience 8040 0.30 <0.001 0.30
Misinformation THREAT Perceived misinfo impact 8040 0.18 <0.001 0.19
Prereg #8 behavioural / participation ABAND Frequency of seeing manipulated news 8040 −0.24 <0.001 −0.26
Prereg #8 behavioural / participation ABAND Non-electoral participation (count) 8040 0.09 <0.001 0.11
Prereg #8 behavioural / participation ABAND Voted in last national election 7912 −0.07 <0.001 −0.08
Prereg #8 behavioural / participation FEAR Frequency of seeing manipulated news 8040 −0.10 <0.001 −0.10
Prereg #8 behavioural / participation FEAR Reactions to suspected misinformation (count) 8040 0.06 <0.001 0.06
Prereg #8 behavioural / participation FEAR Voted in last national election 7912 0.04 0.0024 0.03
Prereg #8 behavioural / participation GEN Frequency of seeing manipulated news 8040 −0.23 <0.001 −0.24
Prereg #8 behavioural / participation GEN Non-electoral participation (count) 8040 0.08 <0.001 0.09
Prereg #8 behavioural / participation GEN Voted in last national election 7912 −0.04 <0.001 −0.06
Prereg #8 behavioural / participation SUPF Frequency of seeing manipulated news 8040 −0.14 <0.001 −0.16
Prereg #8 behavioural / participation SUPF Voted in last national election 7912 −0.07 <0.001 −0.09
Prereg #8 behavioural / participation SUPF Non-electoral participation (count) 8040 0.04 <0.001 0.07
Prereg #8 behavioural / participation SUPFCH Frequency of seeing manipulated news 8040 −0.15 <0.001 −0.16
Prereg #8 behavioural / participation SUPFCH Non-electoral participation (count) 8040 0.06 <0.001 0.07
Prereg #8 behavioural / participation SUPFCH Voted in last national election 7912 −0.02 0.0435 −0.04
Prereg #8 behavioural / participation THREAT Frequency of seeing manipulated news 8040 −0.22 <0.001 −0.25
Prereg #8 behavioural / participation THREAT Non-electoral participation (count) 8040 −0.03 0.0045 −0.02
Prereg #8 behavioural / participation THREAT Voted in last national election 7912 −0.02 0.0759 −0.02
Prereg #8 convergent fairness / helpfulness ABAND Belief others try to be fair 8040 −0.13 <0.001 −0.16
Prereg #8 convergent fairness / helpfulness ABAND Belief others try to be helpful 8040 −0.09 <0.001 −0.13
Prereg #8 convergent fairness / helpfulness FEAR Belief others try to be fair 8040 −0.03 0.0105 −0.03
Prereg #8 convergent fairness / helpfulness FEAR Belief others try to be helpful 8040 −0.03 0.0183 −0.03
Prereg #8 convergent fairness / helpfulness GEN Belief others try to be fair 8040 −0.03 0.0106 −0.07
Prereg #8 convergent fairness / helpfulness GEN Belief others try to be helpful 8040 0.01 0.5716 −0.04
Prereg #8 convergent fairness / helpfulness SUPF Belief others try to be helpful 8040 0.15 <0.001 0.09
Prereg #8 convergent fairness / helpfulness SUPF Belief others try to be fair 8040 0.09 <0.001 0.03
Prereg #8 convergent fairness / helpfulness SUPFCH Belief others try to be helpful 8040 0.17 <0.001 0.12
Prereg #8 convergent fairness / helpfulness SUPFCH Belief others try to be fair 8040 0.12 <0.001 0.07
Prereg #8 convergent fairness / helpfulness THREAT Belief others try to be fair 8040 −0.12 <0.001 −0.14
Prereg #8 convergent fairness / helpfulness THREAT Belief others try to be helpful 8040 −0.08 <0.001 −0.10
Prereg #8 media-channel use ABAND News use — social media / podcasts 8040 0.17 <0.001 0.22
Prereg #8 media-channel use ABAND News use — messaging apps 8040 0.14 <0.001 0.18
Prereg #8 media-channel use ABAND News use — aggregate 8040 0.09 <0.001 0.13
Prereg #8 media-channel use FEAR News use — aggregate 8040 0.19 <0.001 0.19
Prereg #8 media-channel use FEAR News use — TV / radio 8040 0.17 <0.001 0.17
Prereg #8 media-channel use FEAR News use — messaging apps 8040 0.12 <0.001 0.13
Prereg #8 media-channel use GEN News use — social media / podcasts 8040 0.18 <0.001 0.22
Prereg #8 media-channel use GEN News use — messaging apps 8040 0.16 <0.001 0.18
Prereg #8 media-channel use GEN News use — aggregate 8040 0.13 <0.001 0.14
Prereg #8 media-channel use SUPF News use — messaging apps 8040 0.18 <0.001 0.20
Prereg #8 media-channel use SUPF News use — social media / podcasts 8040 0.17 <0.001 0.21
Prereg #8 media-channel use SUPF News use — aggregate 8040 0.12 <0.001 0.14
Prereg #8 media-channel use SUPFCH News use — messaging apps 8040 0.17 <0.001 0.20
Prereg #8 media-channel use SUPFCH News use — aggregate 8040 0.15 <0.001 0.16
Prereg #8 media-channel use SUPFCH News use — social media / podcasts 8040 0.15 <0.001 0.19
Prereg #8 media-channel use THREAT News use — aggregate 8040 0.13 <0.001 0.14
Prereg #8 media-channel use THREAT News use — messaging apps 8040 0.12 <0.001 0.14
Prereg #8 media-channel use THREAT News use — print 8040 0.09 <0.001 0.08
Figure 2: S4 country-adjusted external-anchor correlations. Rows are anchor scales, columns are DisInformeter predictors, and facets separate civic / political from misinformation anchors. Estimates come from outputs/tables/s4_external_anchor_correlations.csv, written by the Study 4 page.

Synthesis

Across the 149 primary preregistered tests, 109 returned a positive verdict (90 confirmed directional or u-shape predictions plus 19 TOST-supported equivalences) and 6 were disconfirmed. The strongest convergent / discriminant signature sits with Exploitation, Abandoned [general], Abandoned by media (censorship) — those constructs absorb the bulk of the confirmed and equivalence-supported counts in the verdict roll-up above. The thinnest external evidence sits with Poisonous ethnocentrism, Superiority China, Fear [general], whose share of confirmed-plus-equivalence-supported tests is the lowest among constructs with at least three preregistered anchors; the 4a measurement architecture page is the appropriate next step for those constructs — weak external evidence often co-occurs with first-order indicators that themselves need short-scale revision. The Fisher-z TOST table separates ‘no significant effect’ from ‘meaningfully small effect’: the equivalence-supported rows are the rows where the prereg’s zero prediction is actively defended by the data, not merely consistent with non-rejection. The S4 anchor panel extends the external-validity picture beyond S5 by adding country-adjusted multinational correlations for civic, political, and misinformation anchors; those estimates are descriptive rather than preregistered verdicts, so they complement rather than replace the S5 hypothesis grid.

References