390 likes | 469 Views
Statistical Methods I & I PSYC 2020 6.0G (F/W 2012). Course Instructor Lisa Fiksenbaum Office: 403 BSB Telephone: (416) 736-5125 E-Mail: lisafix@yorku.ca Office Hour: By Appointment. Teaching Assistant Pearl Gutterman Office: 3023 Lassonde Telephone: (416)736-2100 (ext. 33113 )
E N D
Course Instructor Lisa Fiksenbaum Office: 403 BSB Telephone: (416) 736-5125 E-Mail: lisafix@yorku.ca Office Hour: By Appointment Teaching Assistant Pearl Gutterman Office: 3023 Lassonde Telephone: (416)736-2100 (ext. 33113 ) Email: 2020guterman@gmail.com Office Hour: Tuesday 4:30-5:30 Contact Information
Correspondence by email or phone • Make sure that you identify yourself clearly (first and last name) • Please send emails from a York emailaccount and use PSYC2020 in the subject line; otherwise, emails will be will be deleted unread • Consult the syllabus for administrative information
Book Information Gravetter, F.J. & Wallnau, L.B. (2013) Statistics for the Behavioral Sciences, (9th ed). St. Paul: West Publishing Company.
Rounding • Do not round numbers you are computing until the final answer. • Rounding at each step results in answers that may be significantly different than the keyed answers for both exams and homework. • Round only your final answer (to two decimal places) only after all calculations have been performed.
Course Evaluation • 4 exams (definitions, multiple choice, true/false, matching, basic calculations, interpretation of data sets, and/or short essay questions ): • Exam 1 (20%) • Exam 2 (20%) • Exam 3 (20%) • Exam 4 (20%)
Course Evaluation • 4 assignments: • Assignment 1 (5%) • Assignment 2 (5%) • Assignment 3 (5%) • Assignment 4 (5%) • Due at the START of class (you will receive 0 if handed in late) • NO electronic submissions will be accepted • Do NOT simply report the final answer for a problem. Show the computations that produced that answer.
Exams • For the first exam: • you will be allowed to use one side of a 3 inch x 5 inch index card on which you may put anything you consider useful (e.g., formulas, definitions, etc.). • For all other exams: • you will be allowed to use both sides of the card
Missed Exams • Make-up exams will be granted ONLY under EXCEPTIONAL circumstances, such as serious illness, or death in the immediate family • Must contact the instructor or TA in person, by telephone, or by email, within 48 hours of the missed exam • PROPER DOCUMENTATION REQUIRED
Review of Preliminary Concepts • Variables • Measures of central tendency • Measures of variability • Hypothesis testing
Types of Variables • Variable: characteristics of objects, events, or people that can have different values • Constant: is a characteristic of objects, events, or people that does not vary • Continuous Variable: can take on an infinite number of values (e.g., reaction time) • Discrete Variable: can take on a finite number of values (e.g., gender)
Types of Variables, cont'd • Dependent Variable (DV): the variable being measured in an experiment, that is expected to be “dependent” on the independent variable • Independent Variable (IV): : The variable that is expected to influence the DV • Manipulated IV: an IV controlled by the experimenter (e.g., random assignment to groups) • Subject/Organismic IV: an IV that is an underlying characteristic of the population (e.g., sex, age)
Population/Sample • Population: the entire set of events (e.g., study habits of university students) to which are you are interested • Sample: a subset of a given population that is used to make inferences regarding the population (e.g., an intro psych class)
Parameters/Statistics • Parameter: a measure that refers to the entire population (Greek characters, e.g., µ, , ρ) • Statistic: a measure that refers to a sample (English characters, e.g., X s, r)
Branches of Statistical Methods • Descriptive Statistics: describing the data through frequency distributions, measures of central tendency and variability, etc. • Inferential Statistics: Making inferences about populations by utilizing samples (e.g., are there IQ differences between the sexes)
Measures of Central Tendency (Ch. 3 G & W) • Mean • in the population, this is symbolized by • in the sample, this is symbolized byX • it is calculated by the following formula: X=X N
Mean • Suppose a psychotherapist noted how many sessions her last 10 patients had taken to complete brief therapy with her. The sessions were as follows: 7, 8, 8, 7, 3, 1,6, 9,3, 8 X=X = 60 = 6 N 10
Mean Advantages: • Familiar and intuitively clear to most people • Useful for performing statistical procedures Disadvantages: • May be affected by extreme values • Tedious to compute
Measures of Central Tendency (Ch. 3 G & W) • Median • the score that divides a distribution of scores into the upper and lower halves • aka the 50th percentile • median is better than the mean when there are a few extreme scores
Median • Odd number of scores: line up all scores from lowest to highest, middle score is median 3 ,4 ,5, 7, 8 Median = 5 • Even number of scores: list scores in order (lowest-highest), locate median by finding the point halfway between the middle 2 scores 3, 3 ,4 ,5, 7, 8 Median=4+5 = 4.5 2
Measures of Central Tendency (Ch. 3 G & W) • Mode • most frequently occurring score • may be more than one mode • not affected by extreme values
Mode - Examples • No ModeRaw Data: 10.3 4.9 8.9 11.7 6.3 7.7 • One ModeRaw Data: 6.3 4.9 8.9 6.3 4.94.9 • More Than 1 ModeRaw Data: 21 2828 41 4343
When Do You Use Which Measure? • Categorical or nominal data (e.g., eye colour) - use the mode • Quantitative data (e.g., height, age, test scores) – use the mean and median • Extreme scores - use the median • No extreme scores - use the mean
Central Tendency & Shape of Distribution • Normal Distribution • a purely theoretical distribution • perfectly symmetrical about its mean • Mean=Median=Mode
Central Tendency & Shape of Distribution • Skewed Distributions • Greater proportion of observations fall in one tail of distribution than the other.
Central Tendency & Shape of Distribution • Positively Skewed • tail to right • mode<median<mean
Central Tendency & Shape of Distribution • Negatively Skewed • tail to left • mean<median<mode
Measures of Variability (Ch. 4 G & W) “degree to which scores in a distribution are spread out or clustered” (G& W, p. 104) • Range • difference between the largest and smallest scores in a distribution of scores • isn’t really a good description of the variability for an entire distribution
Measures of Variability (Ch. 4 G & W) • Interquartile Range • difference between the 75th and 25th percentiles in a distribution of scores • the 75th percentile is the score where 75% of scores fall below and the 25th percentile is the score where 25% of the scores fall below
Measures of Variability (Ch. 4 G & W) • Standard Deviation (SD) & Variance • most widely used • determines whether scores are generally near or far from the mean • in the population, the SD is symbolized by and the variance is symbolized by 2 • in the sample, the SD is symbolized by s and the variance is symbolized by s2
Calculating the Variance and/or Standard Deviation Variance: Standard Deviation:
-1 1 6 3 9 10 -2 4 5 -3 9 4 2 4 9 1 1 8 Example: Data: X = {6, 10, 5, 4, 9, 8}; N = 6 Mean: Variance: Standard Deviation: Total: 42 Total: 28
Hypothesis Testing • A hypothesis is a statement about a relationship between variables. The cornerstone of hypothesis testing is the concept of the null hypothesis. • The Null Hypothesis states there is no true difference between scores in the population.
Hypothesis Testing The alternative hypothesis Ha, is that the difference in our sample is truly reflecting a real difference in the population, that the difference is not due to sampling error.
One Tailed Directional hypothesis Eg: “Those receiving $1,000,000 will be happier than the general public” Two-tailed Direction not specified Eg: “Social skills program changes the level of productivity” One –Tailed vs Two-Tailed Hypothesis Tests
Uncertainty & Errors in Hypothesis Testing • Type I error • Null hypothesis is rejected, but it is true • Under control of researcher • α is the probability of making a Type I error
Uncertainty & Errors in Hypothesis Testing • Type II error • Fail to reject null hypothesis when it is false • β is the probability of making a Type II error
Possible Outcomes of Statistical Decision Do not reject Ho Reject Ho Correct Decision Type I Error Ho is True Reality Correct Decision Type II Error Ho is False
Hypothesis Tests in Research Articles (Wang et al, 1997, p. 148)