Changing Rice Bowl: Economic Development and Diet in China: Economic Development and Diet in China

Notes

Notes CHAPTER฀2฀ 1 Th e classic fictional account is Pearl S Buck' s Nobel Prize-winning novel, The Good Earth (1930 ) 2 Overeatin g is also a form o f malnutrition — over-nutritio n CHAPTER฀3฀ 1 Th e first record o f its cultivation i n the United States , where it was regarded a s a botanical curiosity , was in Savannah, Georgia, by Samuel Bowen in 176 5 (Wittwe r et al, 198 7 184 ) CHAPTER฀4฀ 1 Promotio n of agriculture, not manufacturing — the "Food First Hypothesis" — leads to development i n th e form o f better nutritio n Se e Cheng, 198 9 2 Th e Four Modernizations are , in orde r o f priority, Agriculture, Industry, Scienc e and Technology , and th e Militar y CHAPTER฀5฀ 1 A jm i s a litde mor e tha n a pound, 2 jin equa l 1 kilogram CHAPTER฀6฀ 1 Accordin g t o th e Wilcoxo n Signe d Ran k Tes t (use d becaus e o f non-norma l distribution), wit h W = 15 3 0, T + = 203 0, and T - = -50 0 I n othe r words , th e changes i n household s tha t increase d consumptio n totale d 203 , the change s m those that decreased consumption totaled 50, and the total change was 153 Testin g variation b y season doe s not appear in the literature o f foodways and nutritiona l studies Becaus e th e paired- * test examine s change s befor e an d afte r som e intervention t o see whether the intervention had a significant effect, it was chosen for thi s purpos e i n case s where th e dat a were normall y distnbute d Specifically , this test calculates changes resulting from a "treatment", here , the "treatment " is 166฀Notes฀ the change of seasons. Examining the changes, rather than the values themselves, removes th e differences tha t result from individua l responses , producing a mor e sensitive and powerfu l test . The Res t statistic is a ratio of the mean differenc e o f the subjects (household s in this case) before (i n the spring) an d after (i n the fall), divided b y th e standar d erro r o f th e mea n differenc e ( a measur e o f th e approximation with which the mean computed from th e sample approximates th e true populatio n mean) . If the dat a were no t normall y distributed , a Wilcoxon signe d rank s tes t was used i n plac e o f th e paired- * test. Th e Wilcoxo n tes t statisti c W is computed b y ranking al l o f th e difference s befor e an d afte r th e treatmen t (i n thi s case , th e change o f seasons) base d o n thei r absolute value, then attachin g th e signs of th e difference t o th e correspondin g ranks . The signe d rank s ar e the n summe d an d compared. A large ^statisti c indicates that there was a significant chang e becaus e all of th e change s wer e i n on e direction . B y the sam e token , a small W statistic indicates tha t ther e wa s a mixtur e o f positiv e an d negativ e changes , an d the y cancelled eac h othe r out . 2. Th e paired- * test ( t = 2.95) show s a decline. The powe r o f the tes t with alpha a t .05, however, is slightly below the level needed t o give complete confidence t o the results. 3. Usin g the paired- * test ( * =3.52). 4. Th e paired- * test shows * = 2.8; however, the power of that test with alpha .05 was lower than desire d fo r full confidenc e i n th e results. 5. Th e paired-* test for the percentage of the recommended consumptio n of protein was * = 3.26, with a power hig h enoug h t o lend confidenc e t o the results . By the same test , Vitamin C consumption als o declined ( * = 2.13), but th e powe r o f th e test was too lo w to be use d with confidence . Th e chang e...