Cross-sectional measurements of children of the same age may yield spurious growth facts because of great individual differences¶

Claim · All claims
Evidence · 1 study · associational r1 · q1
Currency · newest study 1930 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q1 i? The author argues that supposed growth facts computed from groups of same-aged children may be spurious because variation within a group can obliterate the actual curve of growth. → King 1930

Evidence¶

King 1930¶

King, Irving. (1930). Measurements of the Physical Growth of Two Children. The Journal of Educational Psychology. https://psycnet.apa.org/journal/edu

q1 · i? · associational · r1

This is a theoretical argument from the article's opening discussion, not a tested result. The author reasons that growth inferences from one-time group measurements may be misleading because "the great individual differences" within an age group may mask true growth curves. The article offers no data for this in the argument itself.

"It is possible that all the supposed facts thus brought to light, except those of the most general character, may be quite spurious, on accouut of the great individual differences which may exist in the subjects."

Discussion¶

Related Claims¶

• Group averages are misleading when applied to individual pupils because of extensive overlapping in weight and height across grades — possibly the same claim (merge candidate)
• Correlation magnitude rises with the standard deviation of mental age of the group, so cross-group comparisons of the two traits are spurious — a broader claim this one bears on
• Height at age 12 can be predicted from height at age 6 with a probable error of estimate of 2.98 cm for boys and 2.58 cm for girls — related
• Children maintain their relative height rank within a group from ages 4 to 18, with little crossing of growth curves — related
• Growth measurements at age ten predict stature at sixteen within about 2.5 centimeters and IQ over the same period within 6.3 points — related