click below

click below

Normal Size Small Size show me how

# I/O Psych 542

### Exam 1 Part 4

Question | Answer |
---|---|

THE MEAN, MEDIAN AND MODE OF A NORMAL DISTRIBUTION FALL | AT THE SAME PLACE (VALUE ON A HORIZONTAL AXIS). |

mean | the arithmetic average of a group of scores |

median | the middle score of all the scores in a sample when the scores are arranged in ascending order |

mode | the most common score of all scores in a sample |

guidelines for creating “the Perfect Graph" | •graph has cleartitle? •both axes labeled w/names of variables? all labels read left to right •terms on graph r same terms n textthe graph is for. No abbre.•units of measurement (min., %) •values go down to 0 or hav (double slash) •simple colors •nochartj |

random selection | when we create a sample from a population. almost never used. |

random assignment | refers to a method we can use one we have a sample, whether or not the sample is randomly selected. used frequently and goes a long way when unable to randomly select. |

generalizability | ability for findings from one sample to apply to other samples or contexts. This is external validity because of the fact that you are able to replicate your research and have the same results |

inferential statistics | trying to make general estimates about the population. When we do research, we aren’t able to test the entire population, so we only use a sample. |

standardization | converts raw scores into standard scores so that we are able to compare them with other scores. This way we can make meaningful comparisons |

standardization (continued) | We measure height one way and weight in another; however, with z scores, we can put different variables into the same standardized scale. We can standardize different variables by using their means and standard deviations. This way, we can compare them |

standardization (continued) | RELATE THIS IDEA TO Z SCORES – FOR EXAMPLE, WE CAN COMPARE WEIGHT AND HEIGHT OF A GROUP IN STANDARD DEVIATION UNITS. THE STANDARD DEVIATION UNITS PUT TWO DIFFERENT VARIABLES ON THE SAME MEASURE OR METRIC. |

normal curve is said to be everywhere because | when a pop. is sampled, they tend to fall in a bell curve, which is the normal curve. As size of sample approaches size of the pop. of interest, the shape of the distribution tends to be normally distributed |

parametric tests | statistical analyses based on a set of assumptions about the population. Nonparametric |

non parametric tests | statistical analysis that are not based on a set of assumptions about the population |

statistical power | measure of our ability to reject the null hypothesis, given that the null hypothesis is false. the probability that we will reject the null hypothesis when we should reject the null hypothesis – the probability we will not make a Type II error |

effect size | The size of a difference. A study can have a statistically significant finding; however, it may have a small effect size, which means that while it’s statistically significant, there’s not a large difference |

t test for a single sample | when you have the scores for one sample and want to compare this sample to some known population mean |

t test for dependent means | used when you don’t know a population’s mean or variables and you typically have two sets of sample scores, for instance: pre-test/post-test, repeated-measures design, paired samples (husband/wife, parent/child) |

t test for independent means | you do not know the mean or variance of the population and you have two sets of samples that are completely independent of each other |