At matched token sizes, samples of picture-book text almost never contain fewer unique word types than samples of child-directed speech¶

Claim · All claims
Evidence · 1 study · associational r2 · q2
Currency · newest study 2015 · not yet checked against its sources · not reviewed by a person · overdue for a recheck since 2020: baselines and conditions change

Subclaims¶

q2 i? At a token size of 100, speech exceeded text in unique types in only 9 of 100 paired samples, and at 200 tokens in only 2 of 100; at 300 tokens or greater the ranges were completely non-overlapping. → Montag Jessica L. 2015

Evidence¶

Montag Jessica L. 2015¶

Montag Jessica L., Jones Michael N., Smith Linda B. (2015). The words children hear: Picture books and the statistics for language learning. Psychological Science. https://doi.org/10.1177/0956797615594361

q2 · i? · associational · r2

Results of paired comparisons of 100 book samples with 100 length-matched CHILDES speech samples across token sizes (Figure 1a). The article reports the paired-sample counts quoted and states that in all samples of 300 tokens or greater the ranges of type counts were completely non-overlapping. No effect size is printed.

"at a token size of 100, in only nine of the 100 random samples of speech and text was the number of unique types on speech higher than the number in text and at a token size of 200, in only two of the 100 samples of speech and text were there more unique word types in speech."

Discussion¶

Related Claims¶

• The text of children's picture books contains more unique word types than a length-matched sample of child-directed speech — possibly the same claim (merge candidate)
• Individual picture books typically contain more unique words at a given sample size than individual conversations — possibly the same claim (merge candidate)
• The greater diversity of book input arises mainly because different books sample the language more broadly than different conversations do — a broader claim this one bears on