PSY 101 1st Edition Lecture 5Outline of Last Lecture I. Research Methods and Statistics continueOutline of Current Lecture II. Research Methods and Statistics continuea. Inferential Statistics i. Definition of inferential statisticsb. A foray into frequency distributionsi. Definition of standard deviation ii. Definition of frequency histogram c. Variability d. Statistically significant differencese. The case of Clever Hansf. “Reactivity” and The Hawthorne-Effect g. Many subtle Favors of reactivity i. Definition of subject –expectancy effectii. Definition of Placebo effect iii. Definition of observer-expectancy effect iv. Definition of demand characteristics Current Lecture Inferential Statistics o Inferential statistics: branch of statistics concerned with the confidence with which conclusion about a sample can be generalized to the to population  For example suppose we have two groups of people: men and women.- We examine the means. On average, the men are taller than the women by half an inch.- Is this a significant difference - Can we conclude that, on average, men are taller than women?- Inferential statistic will tell us whether or not the men are significantly taller than women- That is can we generalize our sample results to the population or can it be an errorThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute. A foray into frequency distributionso We use the means and standard deviations of the data Standard deviation: is a measure of the variable of the datao We can examine the distribution of scores, using a frequency histogram  Frequency histogram: a plot of the frequency with which individual scoresappear in our data  Variability o The standard deviation gives us a numerical indication of variability.o Higher standard deviation = grater variability = less uniform datao High standard deviation = all over the graph o Low standard deviation = more uniform data Statistically significant differenceso Inferential stats gives us the likelihood that we will have to obtain this result by chance. If our data are unlikely to occur by chance we deem the difference significant o The p-value is a probability. It tells us how likely our results are to occur due to chance  Calculate using the means and standard deviation of the data o If the p-value is less than .05we have a significant difference. That means the chances of our data occurring due to chance are less than 1 to 20 o Generally speaking, the greater the difference in group means, the smaller the probability that is due to chance.o Conversely, the smaller the variability, the smaller the probability that is due to chance. The case of Clever Hans o Pfungst systematically tested the claims being made by Von Osten Isolated the horse and questioner from the audience. Used a questioner other than Von Osten Used binders to ensure that the horse couldn’t see Von Osten Varied whether the questioner knew the answer to the question  Horse was using body language of owner to find right answer  “Reactivity” and The Hawthorne-Effecto In 1924, Hawthorne Works (and electrical manufacturing company) wanted to increase productivity.o Experimenters manipulated lighting, compensation, number of breaks, and duration of the work day.o Each manipulation increased productivity for a short time, but it quickly decreased again.o Concluded that the changes in productivity were due to increased attention during testing Many subtle flavors of reactivity o Subject-expectancy: Subject expectations about study results influence outcome  Placebo effect: people often recover from a health problem because they expect to get better (similar to a self-fulfilling prophecy) - Discovered by Henry Beecher during World War II when morphinesupplies were running low - Placebos are becoming more affective o Observer- expectancy effect: Experimenter’s knowledge influence participantso Demand Characteristics: Cues in the experiment (or from the experiment) affect how the subject