PSYC 203 1st Edition Lecture 8 Outline of Last Lecture I. Reviewa. Skewness Statistic b. Description of kurtosis c. Correlation coefficients II. Strength or Degree III.Sum of Products IV.Restriction of Range V. Outliers Outline of Current Lecture I. Review II. Restriction of Range III. Golden RuleIV. Causation V. DeterminationVI. Scales of Measurement Current LectureI. Review a. Guidelines about the Size of the Coefficienti. Around +_ .10 is small ii. Around +_ .30 is mediumiii. Around +_ .50 is large These 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.iv. Note that these I conventions proposed by Jacob Cohan and not everyone agrees with them!b. Hand Calculation of the Correlationi. Although it is best to use the formulae given in class, the one given in class works as wellc. Restriction of RangeII. Restriction of Range a. Fact: The obtained value of correlation is affected by the range of scores in the data. b. Ideally, you will have the full range of X and Y observations. c. Otherwise, the sample correlation can be distorted. d. In other words sometimes a sample doesn’t actually represent the actual population for example, if you take a survey that onlyincludes college students you cannot apply those results to individuals that are not college studentse. People often assume that restriction of range makes the observed correlation smaller than the population value. This is not always the case! III. Golden Rulea. Correlation does not prove causation:i. – X could cause Y– ii. Y could cause X–1. Something else could cause both!iii. However, causation implies correlation1. So a failure to find a predicted correlation can beinstructive, sometimes.iv. Big Issue: Third Variable Problem1. For example: if there is a correlation between high coffee intake and low skin cancer, this could be caused By a third variable which would be that most people of that drink coffee on a regular basis work mostly indoorsIV. Causationa. Reverse Causation i. The causal direction is opposite from what has been hypothesizedii. Married people are more likely to be happy. Thus, marriage causes happiness...right?iii. Alternate Explanation: Happy people are more likely to get marriediv. Again, correlation does not mean causationb. Reciprocal Causationi. Two variables cause each otherii. AKA “Spiral effect”/cyclicaliii. Possible Examples:1. Marriage and happiness–a. Being happy selects me into marriage and being married reinforces my happiness2. Coping and depressiona. If you can’t cope you get depressed, if you are depressed you coping skills get hurt3. Volunteering/giving to charity and happinessa. Happy people volunteer and volunteering makes people happyc. Common Causal Variables i. Some unknown third variable is actually influencing boththe variables1. Examples:a. Ice cream sales and drowning rates are positively correlatedb. Third Variable: Summerd. Examplesi. How might you interpret the following results?1. There is a positive correlation between the level of dominance shown by mothers and the level of shyness shown by children.a. It could be that the child is naturally shy andthe parent wants to have fours or encourage them to go out and play with other kids, which seems dominanti. The question becomes is the kid driving the parent or the parent driving the kid?2. There is a positive correlation between loneliness/depression and binge- watching Netflix. a. It could be that depressed people binge watch Netflix or it could be that people that binge watch Netflix get depressedV. Determinationa. the coefficient of determination represents how much variation 2 variables share (in percentages)i. In other words, how much of the variance in variable X can be accounted for by the variance in Y? ii. Basically another way to say how much 2 variables havein common b. The coefficient of determination is found by simply squaring the correlation (r)c. In fact its also called r-squared! d. The coefficient of alienation is what is left over. (1-r^2)VI. Scales of Measurement a. Nominal and Ordinal are categorical b. Interval and ratios are continual c. The difference between first and second in a rank can be very small or a very large this is ordinal you don’t know the actual differenced. In a ratio there is an absolute zero