New User Special Price Expires in

Let's log you in.

Sign in with Facebook


Don't have a StudySoup account? Create one here!


Create a StudySoup account

Be part of our community, it's free to join!

Sign up with Facebook


Create your account
By creating an account you agree to StudySoup's terms and conditions and privacy policy

Already have a StudySoup account? Login here


by: Ariel Kautzer


Ariel Kautzer
GPA 3.83


Almost Ready


These notes were just uploaded, and will be ready to view shortly.

Purchase these notes here, or revisit this page.

Either way, we'll remind you when they're ready :)

Preview These Notes for FREE

Get a free preview of these Notes, just enter your email below.

Unlock Preview
Unlock Preview

Preview these materials now for free

Why put in your email? Get access to more of this material and other relevant free materials for your school

View Preview

About this Document

Class Notes
25 ?




Popular in Course

Popular in Journalism and Mass Communications

This 12 page Class Notes was uploaded by Ariel Kautzer on Tuesday October 13, 2015. The Class Notes belongs to MC 3018 at Louisiana State University taught by Staff in Fall. Since its upload, it has received 25 views. For similar materials see /class/223154/mc-3018-louisiana-state-university in Journalism and Mass Communications at Louisiana State University.

Similar to MC 3018 at LSU

Popular in Journalism and Mass Communications




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: 10/13/15
MC 3018 Test 2 Quantitative Analysis Three Characteristics of Content Analysis 1 It is Systematic a Because you are consistently applying a set of rules that you set up You are going to approach every piece of text the same way Everything you do must be the same 2 Its objective a You are removing your personal observations from the process of analyzing Your observations can inform the creation of the systematic approach but the actual act of collecting data is systematic The ability of anybody being able to use your measure to get the same results It s like universal b That objectivity could challenge notions of signs and culture 3 Its quantitative a You re counting things We want to know how many and how much Why do we use it quantitative content analysis it enumerates puts into numbers qualities of communicative messages and it maps content form and structure When can we use it 1 Describing communication content a Examples i Count the number of violent or sexual instances in prime time television ii Number of female or African American reporters Testing hypothesis regarding message characteristics a If I think that TV sitcoms have greater levels of sexual content now than they did 30 years ago you can test it by looking at prime time b Hypothesis can be about changes in media content 3 Comparing the media to the real world a Comparing how we feel about crime in Baton Rouge to what the media is saying Or we can compare our opinions to actually crime stats Assessing the images it groups in society a Example i Stereotype of stereotypes I want to know how Americans feel towards illegal immigrants You could look at news websites and papers ii You can get a clear picture of how society treats or views a group just by looking at the media 5 You can establish a starting point for a study of media effects a If I want to know how political news coverage affects you as citizens I first have to know what the qualities of political news coverage Iquot P The Process 1 Formulate a research question a Characteristics of research questions should not include anything about people It also has nothing to do with effects of any kind BUT your research question can include questions of categorization time or frequency You are trying to make a numerican picture of text 2 Define the population a If I am doing a content analysis on news coverage of hurricane Katrina I need to know what news coverage I need Just like we do when we sample You would pick through things like TV newspaper blogs podcasts etc We mightjust want Louisiana outlets You could limit your sources 3 Sample a Do we need a sample i How are you going to do it if yes b MultiStage Sampling i Sampling with a purpose Like I m only going to sample the 10 highest circulating papers c Random Sampling i Everyone has an equal chance of sampling d Composite week or month i Looking at a month of news coverage about Katrina create a composite month just out of picking particular days of the week e Purposive i When you chose your sample for the purpose of your study 4 Unit of Analysis a The thing about which you want to make an inference Goes back to your research What or whom you want to make an inference i Ex If you want to know about what kind of commercials are aired during soap operas The unit of analysis would be the commercials 5 Constructing categories a These are the things I want to measure that will tell me about unit of analysis i EX sex on TV b 2 approaches to doing this i A priori Coding constructing categories before you collect data ii Emergent Coding this is your categories or themes emerging while you are collecting the data Three categories of 5 o Mutually exclusive 0 Exhaustive o Reliable 6 Operationalization a You can try to operationalize personal harm i have dead people people who are hurt or something bad 1 Dead specify individual characters a No pulse physical b Verbal confirmation c Zombies 2 Hurt Training and doing a pilot study a You are going to have another person do your coding Your going to take your coding scheme 8 Do the Damn thang a Code it and collect the data 9 Analyize N O Conclusions and ndicaitons a How do we know that this is the case Thursday October 28 2010 Surveys Good They are one of the most frequently used methods in social science research In Political science its almost exclusively what they use They produce a high value of information in a short period of time National Surveys are conducted in a couple of days Cost because you are not doing in a select amount of time they can be cost effective Another advantage to survey research is that it investigates something in a semirealistic setting You can call do surveys in the comfort of your own home They are not constrained by geography There is so much information already out there and chances are that if you have a question about something someone has already done a survey about it Bad You can t manipulate independent variables You are very limited with causation You have correlation but not causation ts getting hard to conduct surveys because people have cell phones With landlines it was easy to get lists of numbers Poor construction If you write or design a survey wrong you can change your results 0 People are shady AKA people are dumb liars It can be a problem using humans as your subjects People have really bad memories People are unpredictable Lastly people s views are very shifting because questions are worded weird What is it It is a systematic method for collecting data for the purposes of constructing quantitative descriptors of attributes of a larger population Statistics Your responses produce statistics They describe characteristics of a population They do it numerically There are two kinds 1 Descriptive a They just describe size and distribution Like what is the mean median and mode What are the percentages Like standard deviations 2 Analytical a They talk about the relationship between two or more variables Steps 1 Research Question a This will always be step one The question should be important related to the real world make a contribution to literature They need to be specific enough to guide the construction of your survey Your research has to describe the population you want to generalize to 2 SamplingDelivery a Sampling is just like we talked about earlier Probability or non probability You have to examine the different generalizations you want to make b How many people do you need to survey c How am I going to administer the survey Look in book They all have there pros and cons Telephone surveys are fast But mailing I might be able to reach more people 3 Questionnaire a It s a lot like developing a coding scheme You have a research question and ask yourself what info I need to know in order to answer my research question Everything on there has to have a purpose Don t include questions you re not going to use 4 Pilot test a When you do a practice or dry run 5 Do the damn thing a Administer your survey until you get the responses you need 6 CleanAnalyze a Make it easy to analyze 7 Report Tuesday November 02 2010 5 Rules for Writing Questions 1 Reflect Goals a It has to reflect the goals of the research project What do you need to know to answer your research question Keep an eye for your results 2 Clear ECT a Your questions should be clear unambiguous and they should accurately communicate what is required of respondents b Use precise phrasing Avoid jargon Don t use slang Don t use crazy acronyms Write the question for the lowest common denominator c Keep the questions short Length is important d Only ask one thing ata time e Avoid biased words or terms This means that you should avoid using things like quotwacky teapartiersquot f Avoid questions with no right answer like llAre you STILL a pothead g Minimize the amount of detail that is required from your respondents h Avoid embarrassing questions i Types of Questions i Open Ended Questions You don t need to provide a finite number of responses They allow you to account for the unforeseeable Not every question falls under a Likert scale question 3 4 5 Assumptions about understanding a Don t assume your respondents will understand what you are asking Pilot testing is super duper important Occams Razor a Simple is better You always want to have an introduction to give a little away You want to have warm up questions The meat of your survey moves from general to specific End with demographic questions Scales etc a Use this to your advantages Tuesday November 09 2010 content analysis IS JUST ABOUT TEXT NOT PEOPLE Experim oldest a ents you can test human behavior and the affect of texts on human behavior It s one of the pproaches to human inquiry We learn through trial and error We have learned most of what we know through experimentation t involv 1 Iquot You can Process 1 es two things Manipulation Changing variables independent We are changing something to see what happens The only reason you do an experiment is so you can manipulate a variable Observation We see if there is a change Pros o Causality like x causes y A change in a variable changes another We cant get this from surveys or focus groups 0 Control our ability to shape not only the environment but the stimuli themselves The more control we have the more internal validity we have 0 Cost pretty cheap 0 Highly replicable you re so precise that it is easy to take another study and do it Cons 0 Selection bias How you go about choosing your participants and signing them to your treatment groups 0 Mortality people not finishing the experiment 0 Testing Affect when your participants are primed to what your experiments about before it happens 0 History Trying to make everyone as same as possible 0 The Hawthorne Effect when people know they are being studied and they change their answers avoid most of these cons Research Question Hypothesis a This is the first time we talk about integrating the hypothesis in They are essential to coming up with an experimental research design You say this is why I think X causes Y They all have to have causality in them b 3 configurations of hypothesis i Comm IV affecting non comm DV 1 EX some kind of media or something impacts a non communication variable like beliefs attitudes or behaviors ii Non Comm IV affects Comm DV 1 This could be Behaviors or beliefs impacting media use iii Comm IV affects Comm DV 2 Setting a Decision between a lab setting and a natural setting b Lab Settings are good when you want to control shit c Natural setting would be the dropping of pamphlets 3 Design a There are four we are going to talk about i DoubleBlind 1 The participants don t know what groups they are assigned to but neither do the researchers a Rare in a MC setting ii PreTest Post Test Control 1 Ask subjects to take a pretest then they are going to be exposed to manipulation then take a post test then we as researches compare the results 2 Random assignment is key Treatment manipulation iii Posttest only control 1 All we are doing is eliminating the pretest 2 Still randomly assign 3 Still have manipulation group 4 Still have post test 5 Experiments where pretest is pointless or where giving a pretest would contaminate results iv Factorials 1 This can test the affect of two or more independent variables on a single dependent variable 2 A factor is the same as one of your independent variables These factors can have multiple levels A level means a degree fa factor 4 Operationalize a Operationalize variables Define what it means and how you are going to measure them You want to talk about independent and dependent separately Focus on the manipulation This is when you decide how many of these levels you want to have 5 Manipulations a This is when you decide on the manipulations themselves There are two kinds of manipulations i Straight Forward 1 Is just where you have a stimuli presented to the participants These are the kind you experience in the MEL ii Staged Manipulation 1 This is where you have events or circumstances that manipulate your independent variable 2 These are a little more elaborate b Your manipulation has to be as strong as possible in order to yield results 6 Subjects a Assign participants to treatment conditions This is where you assign one of the squares to a group Most of the time this is randomly done The effectiveness of experiments is based on random assignment Non random assignment can be okay if your hypothesis calls for it 7 Pilot Study a Make sure were not messing anything up 8 Do the Damn thing a Run it 9 Analyze and report Design and Manipulation Cont d Design You would look at your independed variables to determine your design If you have one your choices are limited If you have 2 or more you need to do a factorial What about my independent variable is important How can I manipulate it Why is it important in real life Tuesday November 16 2010 Variables and Measurements We have to come up with a way to see ifour manipulations did anything 1 We can do it by observation 2 Just ask a Like questionnaire PreTest These are generally where you find independent measures and control variables Sometimes your find primes Examples Regulatory Fit promotion and prevention This is a 2x2 factorial Our pretest is going to be about 11 questions and it is going to tell the researcher whether or not you are more prone to answer to prevention or promotion questions The pretest tests your regulatory fit These are a lot of personality questions The pretest is an independent measure EX If you are a prevention oriented person and you get the promotion questions you would be less likely to vote for the candidate All messages are preventionpromotion LSU Pamphlet One Normal vs CoolAthletic We asked priming questions about how they feel about LSU This helped give us an idea of how they feel We have control variables like what year you are in school or if you plan to go to college Priming is making you use some sort of tool in your brain before you make a decision Like instilling fear Posttests They contain more independent measures dependent measures and controls Independent measures take the form of manipulation checks This checks to see if the manipulation worked Like checking to see if when we primed fear you actually feel fear The dependent measures are the MOST IMPORTANT PART This is the whole reason we are doing the experiment They are going to be some test of whatever your research question is Examples Regulatory Fit We want to know it messages delivered from certain candidates resulted in more positive evaluations Messages are r 39 and r 39 and the quot 39 are male or female Dependent variable is evaluation of the candidate You would need to see how they felt about the message or what they remember or whether they would vote for this candidate Also we could ask whether they thought the candidate was moral knowledgeable or shady You first need to figure out what you need to get at dependent variable then you come up with questions around it Thursday November 18 2010 Making Data Meaningful Once you have a bunch of data you need to figure out what to do with it 1 Clean the Data a Take all of your data and organize it You have to do the thing where you assign your answers numbers Like make the racist questions have high numbers and the less racist lower numbers You have to think of how to make the data logical for use b You would need to give all of the nonresponses neither disagree or agree don t know answers the same code i Get rid of illogical data 1 Like when people transpose number or put that there age is 275 2 Be careful about dates and having to make it to numbers you can understand Statistics They are just mathematical methods that we use to collect organize and summarize data Stats can t correct poor methodology So not having your things well constructed can t be fixed by math 0 Descriptive like baby statistics This goes back to the idea of organizing They describe the size and distribution of various attributes of a population Things like athletes student ect 0 Distribution I These are all of the cases single individual response for a specific variable I FREQUENCY DISTRIBUTION The frequency of these values occurring 0 Like everyone getting 78s 0 Normal Curve I Based on the central limit theorem 0 This says that if you sum together repeated and large numbers of samples and variables and graph the results your will get a normal curve I It says that the majority of responses are going to hover around the average or mean I It allows us to make predictions and calculate error We compare our frequency distribution to the normal curve 0 Summary Stats Measures of Central Tendency they are trying to tell us what the typical score is These take lots of answers and boil them down to a single number There are 3 measures 1 Mean The arithmetic average of a set of scores It takes all of the scores into account The problem with the mean is that it includes the random outliers The extreme values can effect your mean 2 Median The middle value It s the number that falls right in the middle It doesn t account for outliers 3 Mode The most reoccurring number most frequent This gives us only a little bit on information It can actually hind information Measures of Dispersion These tell you about the spread of scores There are 3 1 Range the high versus the low This tells us the distance between the highest and lowest value High minus low 2 Variance The degree to which the scores deviate from the mean We are taking an individual response and subtracting the group mean Then we take that and square it Then we add up all of those and divide by n1 N is the number in your sample There is a problem with the variance The variance comes to us in squared terms This inflates the number were looking at in terms of variance The proportion is off We have to take the square root of our final number This is the standard deviation 3 Standard Deviation the square root of the variance It also talks about the distance from the mean It does so in the original units It has to do with the normal curve We are undoing the squaring of those numbers The reason we squared them is to get rid of negatives a This number is in its original units This is an average of an average We take the average again because we are adding things and then dividing by a number A Z score allows you to compare apples and oranges It is a standardized score that allows you to compare measurements from two or more different methods You take an individual score minus the mean then divide it by the standard deviation Your 2 score is going to always average out to zero So looking at a normal curve the mean is going to be a zero The variance and standard deviation is always going to be one So a z score of 3 is 3 standard deviations away from the mean Tuesday November 30 2010 Hypothesis Testing 0 Criteria for a good hypothesis 0 1 Is compatible with current knowledge I Whatever the latest research or conversation is that our research reflects that o 2 Is logically consistent Syllogistic I Your hypothesis makes sense Like if A then B o 3 Susinct I Short and to the point 0 4 Hypothesis are always testable 2 kinds of hypothesis 0 Null I Basically says nothing is going to happen It says llthere is no effect of impact It is generally the opposite of what you re expecting to happen We are saying any statistical difference between two variables happened by chance or accident I Sampling maybe we took a random sample I Variables may be too closely related 0 Alternative I This is what we actually expect to happen This is saying that the relationship that happened didn t happen by chance That the relationship is not accidental or due to sampling error or measuring error This say that there is an impact and its significant because its real or reality 0 Significance we will develop a test of significance I We re going to set a probability level or significance level We re going to test the null hypothesis against it We care about testing the null Were testing if any of these relationships are by chance If the probability level is lower we can reject the null and accept the alternative Probability level is our way of gambling We set probabilities on the amount of chance were willing to accept 0 When the probability is lt05 were willing to accept a 5 chance we make a wrong decision The lower your prob is the more evidence you need to show significance 0 I Significant results are those that occur on the margin Error there is always a chance that the data you pull supports your alternative hypothesis by accident 0 Type 1 I There is always a chance that the data you pull supports your alternative hypothesis by accident This is when you accept your alternative by accident This is designated by alpha I This is a false positive If a true null means that there is no effect and we find an effect like a pregnancy test You know you re not preggers but your test says you are Type 2 I This is when we accept the null when there actually is an effect When there really is a significant result that we overlook or miss I This is a false negative This is where we are failing to observe an effect when there really is one Your preggers but your test shows your not Type one and type two errors have an inverse relationship When you increase one you decrease the other Like on the curve line Minimizing type one error is the most important I It s because it s an aspect of type two I The false positive is much worse I Like a drug test like seeing a pill treating cancer when it s really not I We are in charge of type one error by setting the significance level 0 O O Final Notes Don t look at o Ch6 o Calculating Reliabilityvalidity formulas 0 Internet Stuff 0 No math or WWW o Ch7 o No computer shit o If it doesn t sound familiar don t fuck with it o If it don t sound fam Don t fuck with it 0 Sample distributions 0 Data transformations 0 Ch 11 0 ALL OF lT is GOOD 0 Ch 12 0 ONLY 2 things we need to know I Basic def ofa Chi square I TTest Short Answers Anything list Guidelines for survey questions Research questions Content analysis research questions NO PEOPLE Expiraments causality Standard DeviationRangeAverage One about expiraments Apply shit It will be a factorial Pros and cons of a Qualitative method Why you use qualitative analysis TF and when to use it Also should be able to know about the mutually exclusive and exhaustive Survey question about income It has to cover all the bases Two approaches to constructing categories Poll Credibility Variance formulas Two kinds of hypothesis basic definitions and how they relate to error Type 1 vs type 2 error 2 kinds of statistics


Buy Material

Are you sure you want to buy this material for

25 Karma

Buy Material

BOOM! Enjoy Your Free Notes!

We've added these Notes to your profile, click here to view them now.


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'

Why people love StudySoup

Bentley McCaw University of Florida

"I was shooting for a perfect 4.0 GPA this semester. Having StudySoup as a study aid was critical to helping me achieve my goal...and I nailed it!"

Amaris Trozzo George Washington University

"I made $350 in just two days after posting my first study guide."

Jim McGreen Ohio University

"Knowing I can count on the Elite Notetaker in my class allows me to focus on what the professor is saying instead of just scribbling notes the whole time and falling behind."

Parker Thompson 500 Startups

"It's a great way for students to improve their educational experience and it seemed like a product that everybody wants, so all the people participating are winning."

Become an Elite Notetaker and start selling your notes online!

Refund Policy


All subscriptions to StudySoup are paid in full at the time of subscribing. To change your credit card information or to cancel your subscription, go to "Edit Settings". All credit card information will be available there. If you should decide to cancel your subscription, it will continue to be valid until the next payment period, as all payments for the current period were made in advance. For special circumstances, please email


StudySoup has more than 1 million course-specific study resources to help students study smarter. If you’re having trouble finding what you’re looking for, our customer support team can help you find what you need! Feel free to contact them here:

Recurring Subscriptions: If you have canceled your recurring subscription on the day of renewal and have not downloaded any documents, you may request a refund by submitting an email to

Satisfaction Guarantee: If you’re not satisfied with your subscription, you can contact us for further help. Contact must be made within 3 business days of your subscription purchase and your refund request will be subject for review.

Please Note: Refunds can never be provided more than 30 days after the initial purchase date regardless of your activity on the site.