A study published in the British Medical Journal (BMJ) in January 2001 tracked a large sample of individuals born in 1946, noting their birthweights and scores on cognitive tests at ages 8, 11, 15, 26, and 43.
a. A 95% confidence interval for difference in mean cognitive scores at age 8 for the low-birthweight group compared with the middle-birthweight group (low minus middle) was (-0.42, -0.11). Explain why this provides evidence that cognitive ability at age 8 is adversely affected by having been a low-birthweight baby.
b. A 95% confidence interval for difference in mean verbal memory scores at age 43 for the low-birthweight group compared with the middle-birthweight group (low minus middle) was (-0.29, +0.09). Explain why this does not provide evidence that verbal memory at age 43 is adversely affected by having been a lowbirthweight baby.
c. Tell which of these, if any, is a possible confounding variable that should be taken into account when examining the relationship between birthweight and cognitive scores: gender, or parents' socio-economic level, or both of these, or neither of these?