Docsity
Docsity

Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity


Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan


Guidelines and tips
Guidelines and tips

EMPIRICAL COMMUNICATION RESEARCH METHODS, Lecture notes of Communication

This course is an introduction to ways of conducting empirical research.

Typology: Lecture notes

2018/2019

Uploaded on 05/18/2019

Kca121
Kca121 🇨🇦

5

(1)

5 documents

1 / 12

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Week 1
Knowledge
- All about agreement
- Authority, tradition, common sense, media, personal experience
Science
- Rationality (logical consistency), reasonableness (open to new ideas), objectivity
- Description (ex: why is the sky blue?), explanation, exploration
Ordinary Inquiry
- Risks of errors associated w non-scholarly methods
-Premature closure: jumping to conclusions
-Halo effect: idea of being influenced by prestige or appearance
Inaccurate Observation
- Casual, semi conscious observation -> mistake (may lose accuracy due to missed
information)
-Solution: scientific observation -> a conscious and deliberate effort to accurately and
precisely observe and measure
Overgeneralization
- Assuming that a few similar events are evidence of a general pattern
- Solution: scientists utilize sufficiently large and representative sample of observation
- Replication of studies
- Can only get a better understanding cuz its normal to be different every time
Selective observation
- Focus on the observations that fit the theory we are using but ignore other sources data
- Solution: social scientists find “deviant cases” (unusual cases)
Illogical reasoning
- Reaching a conclusion through means that are not logical -> Gambler’s Fallacy
- Solution: social scientists make the effort to use systems of logic consciously and
explicitly
Looking for reality
- Scientific assertions must hv both logical and empirical support: logical (the criterion of
assessing the validity of argument)
-Empirical (the criterion requiring sensory experience as evidence)
pf3
pf4
pf5
pf8
pf9
pfa

Partial preview of the text

Download EMPIRICAL COMMUNICATION RESEARCH METHODS and more Lecture notes Communication in PDF only on Docsity!

Week 1 Knowledge

  • All about agreement
  • Authority, tradition, common sense, media, personal experience Science
  • Rationality (logical consistency), reasonableness (open to new ideas), objectivity
  • Description (ex: why is the sky blue?), explanation, exploration Ordinary Inquiry
  • Risks of errors associated w non-scholarly methods
  • Premature closure : jumping to conclusions
  • Halo effect : idea of being influenced by prestige or appearance Inaccurate Observation
  • Casual, semi conscious observation -> mistake (may lose accuracy due to missed information)
  • Solution : scientific observation -> a conscious and deliberate effort to accurately and precisely observe and measure Overgeneralization
  • Assuming that a few similar events are evidence of a general pattern
  • Solution: scientists utilize sufficiently large and representative sample of observation
  • Replication of studies
  • Can only get a better understanding cuz its normal to be different every time Selective observation
  • Focus on the observations that fit the theory we are using but ignore other sources data
  • Solution: social scientists find “deviant cases” (unusual cases) Illogical reasoning
  • Reaching a conclusion through means that are not logical -> Gambler’s Fallacy
  • Solution: social scientists make the effort to use systems of logic consciously and explicitly Looking for reality
  • Scientific assertions must hv both logical and empirical support: logical (the criterion of assessing the validity of argument)
  • Empirical (the criterion requiring sensory experience as evidence)

Human Inquiry

  • Causal reasoning : future circumstances are rooted in or conditioned by present ones
  • Ex: eating a lot without any exercise or activity for long periods leads to obesity
  • Probabilistic reasoning : effects occur more often when the causes occur than when the causes are absent
  • Ex: smoking more likely leads to cancer or texting while drive more likely leads to accident
  • Clinical prediction : often precise predictions offered by experts in the field based on detailed understanding of causes and effects
  • Ex: how many months a patient can live
  • Actuarial prediction : looking for patterns and context in the past and predicting the future
  • Ex: insurance companies credit by the clients usually activities Research focus
  • What problems are worth investigating?
  • What constitutes an answer?
  • Different views on how approaches are grouped Quantitative
  • “‘Numerical” evidence
  • Observations explicit
  • Easier to aggregate, to compare and summarize data
  • Potential for statistical analysis Qualitative data
  • “Non Numerical” evidence
  • Rich in information
  • Captures complex social phenomena well
  • In-depth Some quantitative & qualitative data collection methods
  • See slides Idiographic and nomothetic explanation
  • Idiographic explanation : an accounting that includes multiple causes of a specific event
  • Nomothetic explanation : broad answer Variables
  • Independent (produce changes in dependent variables), ex: education
  • Dependent (changed due to independent variables), ex: income, longevity, success
  • How to define success

Conceptualization & Operationalization Conceptualization

  • Providing names
  • “By which concepts are formed organization of sensory experience” Operationalization
  • Straight forward in an easy way
  • “Process of translating abstract concepts into variables that indicate the concepts” Units of analysis and of observation
  • unit of analysis : the object of a study’s interest (P. 78)
  • Unit of observation : kinds of objects from which evidence is collected Unit of analysis Individual data
  • Evidences gathered about a specific individual Groups
  • In characteristic that belong to one group Organizations
  • Formal social organizations
  • Ex: annual profits, number of students Social artifacts
  • Product of human activity
  • Concrete subject / social interactions Media effects: cultivation theory
  • Into the likelihood
  • Ex: more likely to be influenced by…
  • Possibility for it (not) to happen Distinguishing between media exposure The research question
  • “The guiding premise in the design and analysis of a research study
  • How? -> need to go beyond data The logic causation Nomothetic explanation
  • Centres on variables
  • The dependent variable is the event whose differences are to be explained
  • Causes -> the mechanism or reasons leading to an outcome The variables are correlated
  • Rarely perfect - > how strong a correlation must exist for a connection to be causal
  • Variables do not change tgt systematically -> can’t hv a causal connection The cause occurs before the effect
  • Cause takes place before the effect
  • No causal relationship exists -> cause and effect
  • Ex: education (independent) effects one’s income (dependent) The connection between the variable is non surpious
  • Variables are correlated but not in reality (no fake news)
  • Control variable: identify the context for the relationship between*
  • Wanna see if there is a real connection between the variables Necessary conditions
  • Without it -> doesn’t work
  • Ex: have to be a woman in order to be pregnant Sufficient conditions
  • Hv to try hard in order to succeed in the research Types of hypothesis Null hypothesis
  • Predicts there is no relationship Direct relationship or positive correlation (+&+) inverse or negative correlation (+&-) Way of stating relationships
  • Mostly leads to
  • Is often related to
  • Influences
  • Is associated with... *careful w the usage of word Week 3 Ecological fallacy
  • About the unit as a whole instead of the components of the unit Cross sectional study
  • A study based on observations representing a single point in time
  • Problems: difficult to generalize from one point in time observation

Longitudinal study


Ratio measures : can hv true zero, ex: age Ex: letter grade (A+, B-) -> ordinal Ex: # of Timmies customers in the past 2 months -> ratio (can be 0 number of customer) Ex: Lutheran Anglican, Mormon, Baptist -> nominal Ex: attitudes toward the death sentence -> ordinal Ex: GMAT (some kind of test) scores -> ratio Ex: Population of countries -> ratio Reliability means a particular tool is used to reach similar conclusion

  • The tool is reliable but not accurate Types of reliability
  • Stability: over time
  • Representative: across different subgroups of a population
  • Equivalence: multiple measures can be used to observe same phenomenon
  • Intercoder: researchers use the same measures in making observations, ex: do it today and do it again in 1 month Checking reliability
  • Test retest method
  • Split half method
  • Use established measures: avoid any confusion
  • Reliability of research workers: get feedback from expertise Validity Face & expert panel validity
  • Straight forward, we know for sure its the case Construct validity
  • Just like talking about a hypotheses
  • Ex: satisfied husbands and wives are less likely to cheat on each other Criterion-related validity (predictive validity)
  • Whether particular measure can predict future behaviour correctly
  • Ex: driving test Content validity
  • See slides indicators , indexes, and scales
  • Indicator : an empirical specification of some abstract concept
  • Index : taking into account the number of indicators of the variable, ex: weighting
  • Scale Week 4

Prison experiment

  • Ppl act as a prisoner and the other act prison guard
  • Unethical research
  • Some ppl were feeling uncomfortable Who is vulnerable? (be protected the the research study)
  • Poor ppl, homeless ppl, drug users
  • Basically ppl w less power / sensitive Using statistics Canada Data
  • Identity disclosure -> avoid anything that will expose the interviewee
  • Attribute disclosure -> avoid letting others to hv connections which leads to the interviewee
  • Residual disclosure -> protect interviewee by giving nicknames Contagion: u do whatever others do
  • Ex: ppl smile, u smile back. Ppl in a bad mood, u become sad too
  • Hv to fit in
  • Love <-> love (positive), hate <-> hate (negative) Steps for conducting research in cyberspace
  • have to be voluntary
  • Interviewee can leave anytime they want
  • Leave contact info for the interviewee in case they wanna know more about the research
  • Make sure the interviewees know who you are Tutorial Assignment 1: no book review, has to be empirical research Last sentence of section 1 should mention how is the reading related to the case Annotated bibliography should be after section 1 & 2 Cite the sources if borrow research methods from others Week 5 Nonprobability sampling
  • don’t rely in any random
  • based on reason and logic Reliance on available subjects
  • Use those who can access to -> convenient and inexpensive

Independent variable (stimulus)

  • dichotomous (binomial) -> can be either effective or ineffective
  • Cause & effect
  • The test itself that we r using Dependent variable
  • The impact of the test and we measure Critique
  • Never black or white, always grey Two groups
  • Test 2 groups and compare
  • Control group
  • Don’t watch anything to see if the stimulus has an impact on anyone
  • Experimental group
  • Watch the film Week 8 lab Click the box, analyze -> descriptive statistic -> frequence -> enlarge the chart -> click on a specific question -> move it to the box beside it statistics Mean is numerical average Median is for numerical data Mode is the most responses Sum Charts: bar charts (the best one to use), pie charts How to best represent the chart that is the easiest to communicate? Click on one of the options and select either frequencies or percent -> click ok Click on the chart -> go to edit -> select copy Week 8 Not a numeric variable so there isn’t a mode for statistic Independent variable = smth that cannot change Compare two data: analyze -> descriptive statistic -> crosstabs -> picks for rows and columns (define which one is independent and dependent variable)-> statistics -> chi square -> ok ->

cells -> click observed, expected, row -> click ok -> look at the percentage between 2 variables -> negative vs negative, positive vs positive -> (more or less than 10%?) Chi square less than 0.05 = related Should put interpretation not only for the chat but also the relationship between the two variables Week 9 What is Quantitative Analysis?

  • Write down how many words have u collected
  • Data set -> ex: excel sheet Code book
  • Either we design or borrow from previous study Quantification: The quantified self
  • Ppl pay more attention to numbers Univariate analysis
  • Only focus on thing
  • Dispersion:
  • Range: measure highest to lowest One way distribution
  • ** Contingency table
  • 2 variables Lab / tutorial Put the data in a form that we can use: analyze -> frequency -> statistics -> tick mean, medium, mode -> also tick std. Deviation, minimum, maximum -> continue For the chart, split 0-7, 8-36 instead of 0-8 (see where is the 50% point, closet to 50% but not over 50%) Transform -> record into different variables -> pick the variable that u like -> click ‘old and new values’ -> click range and entre 0 -> enter 7 for through -> click ‘output variables are strings’ -> increase width to 12 or 15 -> go to value and type in ‘lesswork’ (no space) -> click add -> click continue -> punch in new name -> click change -> click continue