Psych Stats SPSS training

Cards

Front Back
SPSS: You are excited—you just found out that you have an IQ of 140. You know something about this test, for instance, that the mean is 100 and the standard deviation is 15. Thus, you know that you are smarter than average, but you decide to do a little math to brag more. What percent of people are you smarter than? This situation uses a score to find a percent of the population, thus you should use a *z-score.*
SPSS: A neuroscientist wants to determine whether the speed of the firing of neurons has anything to do with levels of serotonin in the brain. This situation involves the relationship between two quantitative variables. Thus, a *Pearson correlation* should be run.
SPSS: You received a score of 76 on a stats exam. How did you do? You know the class mean and the standard deviation, but you want to make your raw score more meaningful—you want to know where you fall in the distribution of class score. You want to know the percentage of students who did better or worse than you. You should use a *z-score*
SPSS: Albert works as a risk manager of a large corporation. He has been instructed to determine what factors account for absenteeism at work. He collects data from a sample of employees (age, income, distance from work, stress level, and ratings of health) to try to predict absenteeism. This situation involves the prediction of one quantitative variable from multiple others. Thus, a *Multiple Regression* should be employed.
SPSS: Sarah believes that there is a relationship between academic class ranking in high school and academic class ranking in college. To test this, she samples ten high school students and follows them until they have completed college to get both variables. Is there a relationship between the variables? This question asks for a relationship between two ordinal variables, which requires a *Spearman correlation.*
SPSS: Monique has created a tonic that should increase people’s intelligence. To test its effectiveness, she gives a sample of her assistants the tonic. She then compares them to a known quantity. She knows that the average IQ is 100 with a standard deviation of 15. Did the brain tonic work? this scenario compares a sample to a population. Since both the mean and standard deviation of the population are known, a *one sample z test* should be used.
SPSS: Regina works for the CDC. She believes that she has discovered a disinfectant that kills bacteria for a longer duration on surfaces. She knows that, on average, disinfectants kill bacteria for 2.5 days after being sprayed on any surface. She tests her disinfectant on ten different surfaces then measures how long it takes bacteria to start growing again. Is Regina’s disinfectant more potent that the average disinfectant? This situation compares a sample of scores to a population. Because only the mean is known, a *one-sample t test* should be employed.
SPSS: You want your infant child to know a lot of words. You’ve heard that kids who are read to more frequently know more words on average. You’re not sure whether or not this is true, so you take a sample of two year old children and test how many words they can say. You also ask the parents how many hours they read to the child each week. Does amount of reading time relate to number of words known? This situation asks for the relationship between two quantitative variables, so a *Pearson Correlation* is necessary.
SPSS: I am convinced that humor helps people to be less stressed. I conduct a study in order to demonstrate this hypothesis. I have a group of people who have all taken a questionnaire to measure their level of stress. Then I tell them 10 hilarious jokes and get them all rolling on the floor with laughter before administering the same stress level questionnaire. This situation tests differences between two groups. The design is *repeated measures requiring a related measures* t test.
SPSS: Hugh believes that, through genetic manipulation, he can make children grow taller. He takes a sample of fetuses and changes their genes using radiation. He then follows them until they are 20 years old, at which point he measures their height. He knows that the average adult height is 5.9 feet. Did the radiation make individuals taller? This scenario compares a sample to a population. Since only the population mean is known, a *one sample t test* should be used.
SPSS: You believe that smelling the chemicals from dry-erase markers kill brain cells at a higher rate than normal. You know that, on average, a person loses 100 brain cells every day. In order to test your hypothesis, you find 10 teachers who use dry-erase markers every day and test their level of brain cells at the beginning of the day and then at the end of the day after a normal day of using dry-erase markers. this scenario tests differences in one group on the same test two times. This is a *repeated measures t test* design.
SPSS: You want to determine the effect of gender on math ability.You take a sample of males and females and compare their scores on a math test. Is there a difference between genders in math ability? this situation tests a difference between two unrelated groups, which requires an *independent-samples t test.*
SPSS: You are a track coach who is trying to find a way to find the best long jumpers in the future. You believe that speed has a good deal to do with how far you can jump. You time everyone on your team in the 100 meter dash and divide them into 4 groups: 0-12 sec., 13-15 sec., 16-18 sec., and 18+ sec. After that, you get them all to do the long jump. Are there differences between the groups in how far they can jump? This scenario tests the difference of 1 IV (running speed group) with 3 more levels (4 groups) on 1DV (jumping distance). This requires a *one-way ANOVA.*
SPSS: You are testing the impact of steroid use on baseball players’ ability to hit home runs. You see how many home runs a sample of players can hit in 10 minutes. You then give this sample a steroid injection and repeat the home run contest. You then give them a second injection and repeat the contest. Does steroid injection affect home runs hit? This situation repeats the same measure (hitting home runs). It is a repeated measures design with three different times of testing. Because it identifies differences in one IV (time of home run contest) with three or more levels (and is a repeated measurement) on one DV (number of home runs hit), a *one-way repeated measures ANOVA* would be computed.
SPSS: Curtis is a doctor who believes that some doctors in certain specialties have better general medical knowledge than others. To test this, he takes a sample of cardiologists, pediatricians, oncologists, and podiatrists and has them all complete the test to get into medical school. Does one medical specialty have more general medical knowledge than another? In this example, differences are being tested between four levels of one IV (specialty) on one DV (grad school test). This would require a *one-way ANOVA.*
SPSS: I found a correlation of .40 between IQ and creativity among college students. What test would I use if I wanted to predict a student’s IQ from their score on the creativity scale? This situation asks for a prediction of one variable from another variable when the two variables have a known correlation. A *regression equation* is necessary.
SPSS: I believe that forming a mental image of a word improves memory, but I have no data to support it. To test my theory, I give one group of 20 people a list of 40 nouns and give them 5 minutes to memorize the list. I then ask them to recall as many of the 40 nouns as they can. I give another group of 20 people the same list of words, but in addition to the regular instructions, they are told to form a mental image of each word as they try to recall the list of 40 words. I recorded the number of words recalled for each individual. Can I conclude that mental images affect memory? This situation involves testing the scores from two separate samples to help decide whether there is a significant mean difference between two populations or between two treatment conditions. An *independent-measures t test* is necessary.