TonhauserBeaverDegen2018: experimental results (generated) #
Auto-generated from Linglib/Data/Experiments/TonhauserBeaverDegen2018.json by
scripts/gen_experiments.py. Do not edit by hand: edit the JSON and re-run the generator.
Four rating experiments on the projectivity and at-issueness of 19 projective contents of American English: the content of 9 syntactically heterogeneous expressions (Experiments 1a and 2a) and of the clausal complements of 12 attitude predicates (Experiments 1b and 2b). Experiments 1a and 1b measured projectivity by the certain-that question and at-issueness by the asking-whether question on a no-yes slider, each stimulus a polar question; responses were coded so that 1 is projecting and not-at-issue. Experiments 2a and 2b re-measured at-issueness by an Are-you-sure dissent diagnostic on indicative stimuli. The by-expression means are plotted, not tabulated; the paper prints the mixed-effects estimates, the correlations, two Tukey pairwise-comparison tables and a summary table.
Raw data #
- https://github.com/judith-tonhauser/how-projective: materials, trial-level data and analysis scripts of all four experiments (footnote 8); the by-expression means and the main-clause controls are recomputed from results/exp1a/data/data_preprocessed.csv and results/exp1b/data/data_preprocessed.csv, which already apply the paper's participant exclusions
References #
The four experiments.
- exp1a : Experiment
1a: heterogeneous expressions, asking-whether at-issueness
- exp1b : Experiment
1b: clause-embedding predicates, asking-whether at-issueness
- exp2a : Experiment
2a: heterogeneous expressions, dissent at-issueness
- exp2b : Experiment
2b: clause-embedding predicates, dissent at-issueness
Instances For
Equations
- TonhauserBeaverDegen2018.instDecidableEqExperiment x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
The nine heterogeneous target expressions of Experiments 1a and 2a, with their projective content.
- nrrc : Expression
NRRC: sentence-medial non-restrictive relative clause; its content
- nominalAppositive : Expression
nominal appositive: sentence-medial nominal appositive; the appositive content
- possessiveNP : Expression
possessive NP: possessive noun phrase; the possession implication
- annoyed : Expression
be annoyed: the content of the clausal complement
- discover : Expression
discover: the content of the clausal complement
- know : Expression
know: the content of the clausal complement
- only : Expression
only: the prejacent
- stop : Expression
stop: the pre-state implication
- stupid : Expression
be stupid to: the content of the infinitival complement
Instances For
Equations
- TonhauserBeaverDegen2018.instDecidableEqExpression x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
The twelve clause-embedding predicates of Experiments 1b and 2b; the content is that of the clausal complement.
- annoyed : Predicate
be annoyed: emotive factive
- notice : Predicate
notice: semi-factive
- aware : Predicate
be aware: factive
- realize : Predicate
realize: semi-factive
- amused : Predicate
be amused: emotive factive
- see : Predicate
see: semi-factive
- findOut : Predicate
find out: semi-factive
- learn : Predicate
learn: semi-factive
- discover : Predicate
discover: semi-factive
- reveal : Predicate
reveal: non-factive
- confess : Predicate
confess: non-factive
- establish : Predicate
establish: non-factive
Instances For
Equations
- TonhauserBeaverDegen2018.instDecidableEqPredicate x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
The authors' trigger-class coding in the released data (the classes of Tonhauser et al. 2013).
- b : TriggerClass
B: no strong contextual felicity constraint, no obligatory local effect
- c : TriggerClass
C: no strong contextual felicity constraint, obligatory local effect
Instances For
Equations
- TonhauserBeaverDegen2018.instDecidableEqTriggerClass x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
A printed significance mark; the thresholds differ between tables (*** is .001 in Table 1 and .0001 in Tables 3 and 5).
- threeStars : Significance
***: significant at the table's strictest threshold
- twoStars : Significance
**: significant at .01
- oneStar : Significance
*: significant at .05
- marginal : Significance
.: marginal, at .1
- ns : Significance
n.s.: not significant
Instances For
Equations
- TonhauserBeaverDegen2018.instDecidableEqSignificance x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Whether a printed p-value is an upper or a lower bound.
Instances For
Equations
- TonhauserBeaverDegen2018.instDecidableEqPBound x✝ y✝ = if h : x✝.ctorIdx = y✝.ctorIdx then isTrue ⋯ else isFalse ⋯
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
A row of the by-expression means plotted in the Exp. 1a mean-projectivity-against-mean-not- at-issueness figure (fig:f-proj-ai-1a in the authors' LaTeX); the discover and know means are printed in the summary of Exps. 1a and 1b: the mean certain-that (projectivity) and asking- whether (not-at-issueness) ratings of an expression's content in Experiment 1a, over 210 participants and 17 lexical contents, to two places
- triggerClass : TriggerClass
The authors' trigger-class coding.
- projectivity : Data.Experiments.Decimal
Mean certain-that rating, 1 projecting.
- notAtIssueness : Data.Experiments.Decimal
Mean asking-whether rating, coded 1 not-at-issue.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
The cells of the by-expression means plotted in the Exp. 1a mean-projectivity-against-mean-
not-at-issueness figure (fig:f-proj-ai-1a in the authors' LaTeX); the discover and know means
are printed in the summary of Exps. 1a and 1b, by expression; recomputed from the authors'
released data by scripts/check_experiments.py.
Equations
- One or more equations did not get rendered due to their size.
Instances For
A row of the by-expression means plotted in the Exp. 1b figure (fig:f-proj-ai-1b in the authors' LaTeX); the discover and find out means are printed in the summary of Exps. 1a and 1b: the mean certain-that and asking-whether ratings of a predicate's complement in Experiment 1b, over 235 participants and 20 lexical contents, to two places
- triggerClass : TriggerClass
The authors' trigger-class coding.
- projectivity : Data.Experiments.Decimal
Mean certain-that rating, 1 projecting.
- notAtIssueness : Data.Experiments.Decimal
Mean asking-whether rating, coded 1 not-at-issue.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
The cells of the by-expression means plotted in the Exp. 1b figure (fig:f-proj-ai-1b in the
authors' LaTeX); the discover and find out means are printed in the summary of Exps. 1a and
1b, by predicate; recomputed from the authors' released data by
scripts/check_experiments.py.
Equations
- One or more equations did not get rendered due to their size.
Instances For
A row of plotted with the by-expression means; the asking-whether control mean .02 is printed in the discussion of the at-issueness diagnostics: the mean ratings of the main-clause controls, expected at the floor on both questions
- experiment : Experiment
The experiment.
- projectivity : Data.Experiments.Decimal
Mean certain-that rating.
- notAtIssueness : Data.Experiments.Decimal
Mean asking-whether rating, coded 1 not-at-issue.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
The 2 rows of plotted with the by-expression means; the asking-whether control mean .02 is
printed in the discussion of the at-issueness diagnostics, in the paper's order; recomputed
from the authors' released data by scripts/check_experiments.py.
Equations
- One or more equations did not get rendered due to their size.
Instances For
A row of the regression paragraphs of the four Results sections: the fixed effect of at- issueness on projectivity in the mixed-effects linear regression, with its likelihood-ratio test
- beta : Data.Experiments.Decimal
The estimate.
Its standard error.
The t statistic.
- chiSq : Data.Experiments.Decimal
The likelihood-ratio chi-square on one degree of freedom.
The printed p bound.
- pBound : PBound
Whether p is below or above the printed value.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
The cells of the regression paragraphs of the four Results sections, by experiment; checked against the PDF text layer only.
Equations
- One or more equations did not get rendered due to their size.
Instances For
A row of the at-issueness paragraphs of the four Results sections: Pearson correlations of mean projectivity with mean not-at-issueness, by target expression and without collapsing across lexical contents (at the participant level in Experiments 1)
- byExpression : Data.Experiments.Decimal
r over the by-expression means.
- byItem : Data.Experiments.Decimal
r without collapsing across lexical contents.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
The cells of the at-issueness paragraphs of the four Results sections, by experiment; checked against the PDF text layer only.
Equations
- TonhauserBeaverDegen2018.correlations TonhauserBeaverDegen2018.Experiment.exp1a = { byExpression := { mantissa := 85, exponent := 2 }, byItem := { mantissa := 45, exponent := 2 } }
- TonhauserBeaverDegen2018.correlations TonhauserBeaverDegen2018.Experiment.exp1b = { byExpression := { mantissa := 99, exponent := 2 }, byItem := { mantissa := 44, exponent := 2 } }
- TonhauserBeaverDegen2018.correlations TonhauserBeaverDegen2018.Experiment.exp2a = { byExpression := { mantissa := 84, exponent := 2 }, byItem := { mantissa := 6, exponent := 1 } }
- TonhauserBeaverDegen2018.correlations TonhauserBeaverDegen2018.Experiment.exp2b = { byExpression := { mantissa := 54, exponent := 2 }, byItem := { mantissa := 28, exponent := 2 } }
Instances For
A row of the Participants and Data exclusion paragraphs of each experiment; data points in the regression paragraphs: recruitment, exclusions and the number of data points analysed
- recruited : ℕ
Participants recruited on Mechanical Turk.
- nonNative : ℕ
Excluded as not self-identifying as native speakers of American English.
- outliers : ℕ
Excluded for control means over three standard deviations above the group mean.
- analysed : ℕ
Participants analysed.
- dataPoints : ℕ
Data points in the regression.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
The cells of the Participants and Data exclusion paragraphs of each experiment; data points in the regression paragraphs, by experiment; checked against the PDF text layer only.
Equations
- TonhauserBeaverDegen2018.participants TonhauserBeaverDegen2018.Experiment.exp1a = { recruited := 250, nonNative := 29, outliers := 11, analysed := 210, dataPoints := 1890 }
- TonhauserBeaverDegen2018.participants TonhauserBeaverDegen2018.Experiment.exp1b = { recruited := 250, nonNative := 3, outliers := 12, analysed := 235, dataPoints := 2820 }
- TonhauserBeaverDegen2018.participants TonhauserBeaverDegen2018.Experiment.exp2a = { recruited := 250, nonNative := 6, outliers := 6, analysed := 238, dataPoints := 43 }
- TonhauserBeaverDegen2018.participants TonhauserBeaverDegen2018.Experiment.exp2b = { recruited := 250, nonNative := 6, outliers := 6, analysed := 238, dataPoints := 240 }
Instances For
A row of the Tukey table of the Exp. 1a Results (tab:pairwise in the authors' LaTeX); *** is significance at .001, ** at .01, * at .05, . marginal at .1: Tukey pairwise comparisons of the projectivity means of the Experiment 1a contents; *** at .001
- first : Expression
The row expression.
- second : Expression
The column expression.
- significance : Significance
The printed mark.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
The 36 rows of the Tukey table of the Exp. 1a Results (tab:pairwise in the authors' LaTeX); *** is significance at .001, ** at .01, * at .05, . marginal at .1, in the paper's order; checked against the PDF text layer only.
Equations
- One or more equations did not get rendered due to their size.
Instances For
A row of the Tukey table of the Exp. 1b Results (tab:pairwise-1b in the authors' LaTeX); *** is significance at .0001, ** at .01, * at .05, . marginal at .1: Tukey pairwise comparisons of the projectivity means of the Experiment 1b complements; *** at .0001
- first : Predicate
The row predicate.
- second : Predicate
The column predicate.
- significance : Significance
The printed mark.
Instances For
Equations
- One or more equations did not get rendered due to their size.
Instances For
Equations
Equations
- One or more equations did not get rendered due to their size.
Instances For
The 66 rows of the Tukey table of the Exp. 1b Results (tab:pairwise-1b in the authors' LaTeX); *** is significance at .0001, ** at .01, * at .05, . marginal at .1, in the paper's order; checked against the PDF text layer only.
Equations
- One or more equations did not get rendered due to their size.