Honors Statistics Aug 23-8:26 PM Daily Agenda 1. Check homework C4#9 Finish 4.2 concepts 4. Discuss 4.3 concepts Aug 23-8:31 PM 1
Apr 6-9:53 AM Nov 11-12:33 PM 2
Lack of BLINDING... The same person "experimenter" that ranked the subjects anxiety levels at the beginning of the study also taught the meditation techniques and then did the final rating. There should have been a different person doing the ratings (one who is not told which treatments the subjects were given) There is too much pressure to provide successful results when you are designing, rating and implementing the experiment. Nov 9-2:05 PM testosterone? Double Blind means that neither the patients who were taking the medicine nor the doctors evaluating the patients knew what would be needed to keep track of the medicines and treatment Randomized. This refers to the practice of selecting the patients for the treatment groups by a method that involves chance. This will provide for roughly equivalent treatment groups that will distribute the other possible confounding variables throughout the treatment groups. Placebo Controlled. This refers to the practice of providing all patients and treatment groups with a "treatment". Everyone involved in the experiment got something to take. This will separate the placebo effect from the active treatment medicine. The factor involved when a person takes a pill has an expectation for healing to occur because of begin given a treatment. Nov 4-7:57 PM 3
Acupuncture and pregnancy A study sought to determine whether the ancient Chinese art of acupuncture could help infertile women become pregnant.43 One hundred sixty healthy women undergoing assisted reproductive therapy were recruited for the study. Half of the subjects were randomly assigned to receive acupuncture treatment 25 minutes before embryo transfer and again 25 minutes after the transfer. The remaining 80 subjects were instructed to lie still for 25 minutes after the embryo transfer. Results: In the acupuncture group, 34 women became pregnant. In the control group, 21 women became (a) Why did researchers randomly assign the subjects to the two treatments? Randomization reduces bias and creates roughly equivalent treatment groups. This provides for all other possible confounding variables to be distributed throughout the treatment groups before treatment begins. (b) The difference in the percent of women who became pregnant in the two groups is statistically significant. Explain what this means to someone who knows little statistics. The treatment results were so largely different that they would not have happened very often by just chance alone. We consider the result to "mean" that the treatment worked. The acupuncture provided a positive outcome, that of becoming (c) Explain why the design of the study prevents us from concluding that acupuncture caused the difference in pregnancy rates. While there was a comparison group, this group did not receive any "treatment" so it is hard to conclude a cause and effect relationship in this study. Perhaps the placebo effect occured because the first group of women received an actual treatment. Nov 4-7:58 PM Oct 27-11:38 AM 4
Oct 27-11:26 AM Oct 31-11:19 AM 5
Nov 9-2:03 PM Nov 10-3:50 PM 6
Nov 9-2:04 PM Nov 9-1:58 PM 7
The blocks are the diagnoses... we can not randomly assign chronic health conditions to patients... A randomized block design is preferable because we seek to determine if nurse-practitioners are a capable as doctors in treating the patients. Perhaps the care of certain chronic conditions require more expertise than others. If we think that the type of health condition makes a difference in the type of health care needed we should "control" for this factor in our experiment before we begin to decrease the variability in the C) Single blind, double-blind or no blind? D) Create flowchart for experiment. Nov 4-6:54 PM MATCHED-PAIRS DESIGN. Nov 9-2:04 PM 8
It is an experiment because we have a Squeezing the scale What "handedness" you are, right handed people are stronger with their Force exerted when squeezing the scale. Units are most likely pounds or kg. Apr 3-1:57 PM Squeeze right hand then left hand record force exerted for both Apr 3-1:58 PM 9
1 10 1 group 1 (1,2,4,8,5) group 2 (3,6,7,9,10) Apr 3-1:58 PM a) The treatment groups are confounded by gender. If a difference is found the students will not know if it is due to the deodorant brand or the gender difference. Each group should have both treatments. An even more effective design... Matched Pairs Brand A right pit Brand B left pit Brand A left pit Brand B right pit Nov 4-6:32 PM 10
Nov 9-7:08 AM Nov 9-7:08 AM 11
Nov 9-3:20 PM Nov 9-3:20 PM 12
Show cause and effect generalize results to the population Nov 9-3:21 PM Random sampling versus random assignment Explain the difference between the types of inference that can be made as a result of random sampling and random assignment. Random sampling - Generalize results to the population of interest. Random assignment - Show (prove) a cause and effect relationship Causation. Nov 9-3:27 PM 13
Observation versus experimentation Explain the difference between the types of inference than can usually be made from an observational study and an experiment. Observational Study Show that an association exists Experiment Study Show (prove) a cause and effect relationship Causation. Nov 9-3:28 PM Foster care versus orphanages: Do abandoned children placed in foster homes do better than similar children placed in an institution? The Bucharest Early Intervention Project found that the answer is a clear Yes. The subjects were 136 young children abandoned at birth and living in orphanages in Bucharest, Romania. Half of the children, chosen at random, were placed in foster homes. The other half remained in the orphanages. 55 (Foster care was not easily available in Romania at the time and so was paid for by the study.) What conclusion can we draw from this study? Explain. Random Selection - No random selection from any population, children were abandoned at birth and living in Bucharest orphanages. We cannot generalize the results to other orphans. Random assignment - Yes children were randomly assigned to treatment groups.this would allows us to conclude (infer) that children placed in foster care "do better". However... Bad experimental design... so... Nov 9-3:18 PM 14
Group Work Nov 1-3:27 PM Frozen batteries Will storing batteries in a freezer make them last longer? To find out, a company that produces batteries takes a random sample of 100 AA batteries from its warehouse. The company statistician randomly assigns 50 batteries to be stored in the freezer and the other 50 to be stored at room temperature for 3 years. At the end of that time period, each battery s charge is tested. Result: Batteries stored in the freezer had a higher average charge, and the difference between the groups was statistically significant. What conclusion can we draw from this study? Explain. Random selection - This allows us to (infer) that the experiment results will generalize to the population of AA batteries of this type at this warehouse. Random assignment - This allows us to conclude (infer) that placing batteries in the freezer causes them to last longer. Nov 9-3:30 PM 15
Who talks more women or men? According to pg 267 Louann Brizendine, author of The Female Brain, women say nearly three times as many words per day as men. Skeptical researchers devised a study to test this claim. They used electronic devices to record the talking patterns of 396 university students who volunteered to participate in the study. The device was programmed to record 30 seconds of sound every 12.5 minutes without the carrier s knowledge. According to a published report of the study in Scientific American, Men showed a slightly wider variability in words utteredf. But in the end, the sexes came out just about even in the daily averages: women at 16,215 words and men at 15,669. 56 This difference was not statistically significant. What conclusion can we draw from this study? Explain. Random selection - There was no random selection... volunteers agreed to wear the electronic devices. We cannot generalize the results to any population. Random assignment - There was no random assignment and no actual treatment. This was an observational study. No cause and effect can be concluded. Nov 9-3:36 PM Attend church, live longer? One of the better studies of the effect of regular attendance at religious services gathered data from a random sample of 3617 adults. The researchers then measured lots of variables, not just the explanatory variable (religious activities) and the response variable (length of life). A news article said: Churchgoers were more likely to be nonsmokers, physically active, and at their right weight. But even after health behaviors were taken into account, those not attending religious services regularly still were about 25% more likely to have died (earlier than expected.) 57 What conclusion can we draw from this study? Explain. Random selection - This allows us to (infer) that the result (longer life) will generalize to the population adults from which the sample was selected. Random assignment - This was not an experiment. No random assignment so no cause and effect can be inferred. Nov 9-3:38 PM 16
pg 262-264: 76, 79, 81, 86 TEAM QUIZ TOMORROW: look over baby chicks worksheet Apr 6-9:53 AM Mar 28-9:20 AM 17