MATH 109 Lecture Notes - Lecture 3: University Of California, Berkeley, Scientific Control
Document Summary
Controlled experiment as well as observational studies are good methods for collecting data. However, observational studies are not as successful as controlled experiment. In a controlled experiment, the investigator decides who will be in the control group and who will be in the treatment group. In an observational study, the subjects assign themselves to one of these two groups: the studies on. The effect of spanking a child (can lower iq. Are some examples of observational studies: in any observational study, confounding is a major source of bias, to avoid the effects of hidden confounding factors statisticians make comparisons separately for smaller and more homogenous groups. Compare gender to gender or age to age: in the uc, berkeley example, we have seen that although the graduate school admitted. 44% male applicants and 30% female applicants, the admission rate was because of the choice and not because of sex discrimination. Women decided to take harder majors than man.
