Comm106 Midterm Study Guide
Comm106 Midterm Study Guide Comm106
Popular in Communication Research Methods
Popular in Communication Studies
verified elite notetaker
This 6 page Study Guide was uploaded by Erica Evans on Monday February 1, 2016. The Study Guide belongs to Comm106 at Stanford University taught by Jennifer Pan in Fall 2016. Since its upload, it has received 82 views. For similar materials see Communication Research Methods in Communication Studies at Stanford University.
Reviews for Comm106 Midterm Study Guide
Yes YES!! Thank you for these. I'm such a bad notetaker :/ will definitely be looking forward to these
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/01/16
Comm106 Midterm Study Guide 2/1/2016 Terms: 1. Theory: A conjecture about the explanation of some phenomena of interest. Aka: an idea that explains a thing! -‐ Observable implications -‐ Falsifiable -‐ Practical way to test it -‐ As general as possible -‐ As parsimonious as possible Quantitative data: Numerical measures of things Qualitative data: Deep description and interpretation; gathered through case studies and interviews; captures subtleties of meaning Concepts: ideas from the real world that we are theorizing about Conceptual definition: Explains how you measure a characteristic. Template: The concept of __________ is defined as the extent to which __________ exhibit the characteristic of _________. Unit of analysis: the level at which you are measuring a thing Operational definition: How you will do the testing for the concept. Ex: To test for happiness we will ask people to rate how satisfied they are with their lives on a scale of one to ten. Measurement error: inaccuracy in your data because the test does not perfectly capture reality. Systematic measurement error: When there is something intrinsically wrong with the experiment such that it consistently produces error. Valid measure: When there is no systematic measurement error. Random measurement error: This exists most of the time and is due to conditions you cannot control. Reliable measure: When there is no random measurement error. Face validity: The extent to which an experiment seems like a reasonable measure of a characteristic. Nominal variable: When you can separate the subject into categories but cannot order them: like hair color. Ordinal variable: When you can separate the subject into categories and order them, but they are not quantifiable: like always, sometimes, rarely, never Interval-‐level variable: When you can continuously measure the subject and know exact differences: like weight 2. Spurious association: When there is correlation but not causation. X and Y may demonstrate a relationship, but a third variable Z is the real cause. Post hoc fallacy: In order for X to cause Y, X must come before Y. But X coming before Y is not enough to prove it causes Y. No association: The fact that there is no association does not mean there is no causation: For example there could be some selection mechanism that skews the data. Counterfactual: what would have happened if X had been different in some way Causal effect = actual – counterfactual: The difference between what actually happened and what would have happened in the counterfactual case. Fundamental problem of causal inference: We can’t go back in time! Controlled experiment: Involves finding subjects, avoiding sample selection bias, randomly assigning some subjects to treatment group and some to control group, and then measuring outcomes before and after the experiment. Field experiment: An experiment that takes place in a subject’s natural environment and not in a lab. Lab experiment: Takes place in a lab. You can easily manipulate the treatment and create environments that are not available in real life. Internet experiment: Similar to a field experiment, but it happens online. Natural experiment: When nature or the government performs the assignment for you. Example: the Vietnam War draft, which was based on lottery. Observational study: When you cannot control the assignment of X, so the assignment is not random, but you can observe it. External validity: The extent to which your results are generalizable to other populations Internal validity: The extent that you know your results are due to your manipulation only: how sure you are that X is driving Y. Construct validity: The extent to which you are actually measuring what you think you are. 3. Descriptive statistics: Various ways to summarize data. Frequency tables: Used with nominal variables – tells you how many among the data are in each category. Valid percent: A percentage that excludes the people that didn’t respond. Measure of central tendency: what is a typical or average value Mode: The most frequent value (the only valuable measure for nominal variables because you can’t rank them) Bi-‐modal distribution: When there are two categories and they have the same distribution. Median: The value in the middle when we rank all values of the variable (can be used for ordinal variables) Quantiles: When you divide observations into groups. Quartiles: When observations are divided into 4 groups. Inter-‐quartile range: The difference between the third and first quartiles (so basically the middle section; this helps remove outliers. Percentile: When you divide the observations into 100 groups. Mean: The sum of all observations divided by the total number of observations (can be used for continuous variables) Histogram: Each bar in this table represents the values of certain intervals, called “bins.” The taller the bar, the more values in that bin. Range: The difference between the largest and smallest numbers in the sample. Outlier: when an observation falls more than 1.5 x IQR above the 3 quartile, or the rd observation falls 1.5 x IQR below the 1 quartile. st st rd Boxplots (or box and whisker plots): This table shows the median, and 1 and 3 quartiles. Everything beyond the “whiskers” is an outlier. Standard deviation: On average, how far away are the data points from the sample mean? 4. Population: The universe of cases we want to describe. Population parameter: the characteristic of the population we care about. Ex: What is the average income of adults in the U.S.? The population parameter is average income. Census: When we have every single case documented. But this is really expensive! Sample: A selected subset of the population of all the cases. Sample statistic: The word used to denote the population parameter in a sample case – i.e. what we use to “estimate” the population parameter. Statistical inference: Saying something about the population from the sample. Inference: When you infer something about the world beyond what we observe – this is what makes research scientific. Simple random sampling: Every member of the population must have an equal chance of being chosen for the study. Sampling bias: If you have the wrong sampling frame. Random sampling error: When your sample data differs from the actual data by a random error. The more variation in the population, the more random sampling error there will be Working with R: R script: The document where you write your code that can be saved. Console: The space where the result of your code appears when you run it. This does not get saved. Objects: • R stores information as objects. Once you make objects, you want to refer to them by that name. • <-‐ is the assignment operator • You can assign values to certain objects • Or a string of characters using quotation marks • If you reassign a new value to the same object name, the original object will be overwritten • Object names are case sensitive, cannot start with a number, but they can include a number, cannot include spaces, can include periods or underscores Class: The type of object; is it a number, string (like a word or group of words) or function? Functions: Performs some action on an object Vectors: One dimensional lists of information. Combine elements in a vetor with the c() function. Ex: vector.name <-‐ c( “element”, “element2”, “element3”, “etc”) Arithmetic: • 5+5, 5-‐5, 5*5, 5/5, sqrt(5), log(5) etc. • In R, spacing doesn’t matter • Order of operations applies Loading data: 1) Set working directory à setwd("/Users/Erica/Downloads") 2) Read in the data and assign to an object à afghan.class <-‐ read.csv("afghan-‐ village.csv") Sub-‐setting: • Use these symbols to narrow down your variables: == (equals) < (less than) > (more than) <= (less than or equal to) >= (more than or equal to) != (does not equal) • Ex: afghan.subset <-‐ afghan[afghan$age <= 30, ] #leave a space to include all the columns • Using the subset function: The subset function has three arguments 1) first you tell it the data set 2) the second argument is “select =” which selects the specific columns. You need the c() if you are selecting more than one column. The third argument is “subset =” and this selects the rows you want to include. • Example: • afghan.li <-‐ subset(afghan, select = c("violent.exp.ISAF", "violent.exp.taliban"), subset = (income == "less than 2,000") | (income == "2,001-‐10,000") | (income == "10,001-‐20,000")) Descriptive statistics: • Median à median() • Quantiles à summary() • Range à range() • Interquartile range à IQR() • Standard deviation à sd() • Variance à var() Tables: • Frequency table à table() • Bar chart à barplot() • Histogram à hist() • Boxplot à boxplot()
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'