Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
LINB29 Lab
University of Toronto Scarborough Answer the following questions. You should use R to answer the questions, and you may consult any of the resources we’ve used in class (lecture slides, your notes, the R guide on Quercus), but you should not discuss the lab with other students in the class. The data for the questions is available on Quercus in Week 7, along with this instruction sheet. As usual, submit the write-up of your answers on Quercus. Office hours: I’ll be in the Zoom room for our classroom from 13:00 to 15:00 Wednesday, 10/27, in order to answer any questions about the lab. Office hours will be held as regular that week. Due date: The write-up for this is due 10/31/2021, 23:59 Toronto time. 1 The length of silence (19 points) A previous report found that the mean duration of silence as a filler in speech is 716ms in Canadian English. You have reason to believe that speakers from Toronto have longer silences than has been reported for Canadian English. You do your own data collection to see whether this is true and measure the length of silence from thirty speakers from Toronto. Use the data in labq1.txt to determine whether silence in Toronto is longer than silence in Canadian English in general. 1. Compute the mean, median, standard deviation, upper 95% confidence interval, and lower 95% confidence interval for silence. (5 points) 2. State and motivate your choice of statistical test to make the relevant comparison. Make sure you report any additional tests you ran to meet the assumptions of the test. (4 points) 3. State your hypotheses for the test you’ve chosen. (2 points) 4. Given the 95% confidence interval, can you make any inferences about the result before running the statistical test? Why or why not? (3 points) 5. Report and interpret your results. (5 points) 1 2 Lexical decision reaction times (17 points) You decide to run a lexical decision task experiment in order to compare the reaction times for low frequency and high frequency words. You test thirty-five speakers and gather information on each’s reaction time for both low and high frequencywords (the data is saved in labq2.txt). Are the reaction times for low frequency and high frequency words for these speakers different or the same? 1. Compute the mean, median, and standard deviation of the reaction time for each level of the independent variable. (6 points) 2. State and motivate your choice of statistical test. Make sure you report any additional tests you ran to check the assumptions of the test. (4 points) 3. State your hypotheses. (2 points) 4. Report and interpret your results. (5 points) 2 3 Referential Violations (10 points) A psycholinguist examined acceptability ratings for different Referential Violations. In par- ticular, sentences with Principle A and Principle B violations, such as in (1) and (2). Thirty participants rated two sentences each on a Likert scale from 1 to 5 (1 is “completely unaccept- able”; 5 is “completely acceptable”). Note: You do not need to use R to answer this question. (1) *Maryi thought that John saw herselfi. (Principle A violation) (2) *Mary thought that Johni saw himi. (Principle B violation) The psycholinguist reports, “The mean rating for Principle A violations (mean: 1.4) was lower than Principle B violations (mean: 2.2). This difference was significant, based on a one-tailed paired t-test (t = 2.64, df = 29, p < .05).” 1. What is the independent variable? What are its levels? What type of variable is it? (3 points) 2. What is the dependent variable? What are its levels? What type of variable is it? (3 points) 3. Is a t-test an appropriate test to use for data of this sort? Why or why not? If a t-test is not appropriate, which test would be more appropriate? (4 points) 3 4 Pronoun comprehension and Down syndrome (17 points) Researchers tested whether children with Down syndrome have deficits in understanding pro- nouns, compared to typically developing children. Fourteen Green-speaking children with Down syndrome (DS) and fourteen typically developing children (TD) were recruited from schools. They listened to sentences involving a pronoun like the one in (3), and completed a picture-matching task where they selected which picture correctly matched the sentence. (3) O babas the dad skupizi is drying ton eafto tu himself (Greek) “Dad is drying himself.” Compare whether the TD children and the DS children comprehend reflexive pronouns in similar or different ways. The REFLEXIVE column in the data records the number of correct answers to the picture-matching task for reflexive sentences, and the GROUP column states whether the child was TD or had DS. The relevant data is in labq4.txt. 1. State and motivate your choice of statistical test. Make sure you also report any addi- tional tests you ran to meet the assumptions of the test. (4 points) 2. State your hypotheses. (2 points) 3. Compute mean, median, and standard deviation for TD and DS children. (6 points) 4. Report and interpret your results. (5 points)