Ch. 1: Intro to Statistics and Research Design
A researcher recruits 100 students from the entryway of a library to examine gender differences in book genre consumption. The researcher is employing this type of research study design.
What is a correlational design? Will also accept between-groups design
This type of chart includes two (or three) columns -- a label for each value, and a count of how many times that value occurred.
What is a frequency table?
This type of graph visually depicts the relationship between two scale (a.k.a. interval or ratio) variables by plotting each individual point, with the independent variable on the x axis and the dependent variable on the y axis.
What is a scatter plot?
These are the three measures of central tendency, ordered from least to most affected by outliers
What are the mode, median, and mean?
(mode will not change at all after introducing an outlier; median will have little to no change, mean will have a large change)
These are 1 example for each type of scale: nominal, ordinal, interval, and ratio
What are...
nominal: name, ID, group label, address, phone number, etc.
ordinal: class rank, top 5 favorite movies, vote in a presidential election (arguable?)
interval: temperature (in degrees Fahrenheit or Celsius), number of blinks in an hour (maybe?), number of days in a week
ratio: age, height, weight, number of apples eaten in a week, amount of time before pressing a button, etc.
Will accept others.
A researcher randomly selects 100 students from the BU directory to participate in a research study. 50 students are asked to read an excerpt from a horror novel; the other 50 are asked to read from a romance novel. After reading, participants are shown a short anxiety-provoking video and asked to rate their level of discomfort. The researcher is employing this kind of study design.
What is an experimental design? (Also a between-groups design)
This type of graph visually depicts frequencies for quantitative variables over a range (with no gaps between bars).
What is a histogram?
A researcher randomly selects 1000 BU students for a research study on religious affiliation. This type of chart would best depict the relative proportions of the groups assessed.
What is a pie chart? (Bar chart OK, but note the difference between "frequencies" and "proportions")
These are three measures of variability.
What are range, standard deviation, variance? Will also accept interquartile range (difference in value between 75th percentile and 25th percentile)
This is the major distinction between a correlational vs an experimental research design
[Daily double] What is experimenter control? (The researcher must be able to separate into an active group who "receive a treatment" & a control group who "receive nothing") (Nuance can be added here..., e.g. placebo effect)
A researcher randomly selects 100 students from the BU directory to participate in a research study. 50 students are asked to read an excerpt from a horror novel; the other 50 are asked to read from a romance novel. After reading, participants are shown a short anxiety-provoking video and asked to rate their level of discomfort. What is (probably) the researcher's hypothesis?
What is Higher movie-induced discomfort ratings in the horror-readers than the romance-readers? (One-tailed hypothesis)
Will also accept Difference in discomfort ratings between the two groups (Two-tailed hypothesis)
This type of graph visually depicts frequencies for categorical (a.k.a nominal, a.k.a qualitative) variables (with gaps between bars).
What is a bar chart?
These are 3 key components of a graph needed in order to effectively interpret any visualization of data.
What are the title, axis labels, axis values (or "tick labels")? Will also accept key or legend (for categorical data).
These are (at least) two properties of a distribution which are not fully captured by measures of central tendency or variability.
What are skew and kurtosis? (Will accept others)
Suppose 500 students are enrolled in PS 2111, an imaginary class which deals with very advanced statistics and research design. At the beginning of class 50 students are selected to take a quiz to assess incoming students' familiarity with different topics. In this design, the 500 students are the _________, and the 50 students are the _______.
What are the population and the sample?
A researcher recruits 100 students from the entryway of a library to examine gender differences in book genre consumption. The researcher is employing this type of sampling procedure.
What is convenience sampling?
This type of distribution is symmetric and unimodal -- but it doesn't chime!
What is the normal distribution? (AKA the bell curve)
These are (at least) 3 components of a graph one can use to distinguish group differences within a plot
What are color, size, shape, error bars, significance markers, etc.?
This measure of variance can be used as a cursory tool for detecting outliers, as it ignores influence from extremely low or extremely high values
What is the interquartile range?
(P.S. 1.5 * IQR is a common method of outlier detection)
Continuous data may be drawn from one of these two scale types.
What are interval and ratio scales? (A continuous variable can never be nominal or ordinal)
Researchers must carefully define variables in order to accurately capture relevant information. This process occurs when a researcher explicitly states how they plan to measure a variable they are interested in.
[Daily double] What is operationalization? (AKA making an Operational Definition)
This is a 2-part description of the shape of the following data:
1,2,3,3,3,3,4,6,7,8,8,8,8,9,11,17
What is a bimodal distribution with positive skew?
These are (at least) two ways to lie with a data visualization
What are wonky axes (discontinuous, inconsistent or flipped), exclusion of variability information (i.e. no error bars or significance markers), purposeful misinterpretation of results? (Will accept others)
Imagine you are a researcher facing a dreaded experience with data corruption - luckily, you were able to compute some statistics before the corruption occurred. Suppose you have the following information saved:
n = 4
minimum = 1
mean = 5
median = 4
mode = 4
These values correspond to your missing data.
What are 1,4,4, and 11?
In experimental research, this type of variable is manipulated by the researcher to observe its effect on other variables
What is the independent variable?