MCOM 390 Exam two

Pray

Exploritory purpose

\
* Researchers conducting exploratory studies are interested in investigating, exploring, or attempting to figure out a new, innovative thread of knowledge. 
* Can be either qualitative or quantitative

Descriptive purpose

\
* Allows researchers to describe a phenomenon or understand the details about people's experiences of a particular event
* Generally qualitative because its often necessary to to gather in-depth data from people and allow participants the flexibility of expressing their feelings and situations in detail

Explanatory Purpose

\
* Focuses on explaining the reasons behind a phenomenon, relationship, or event
* Seeks to explain the reasons behind an already known and/or established relationship between
* Explanatory studies can be qualitative or quantitative it depends on what the researcher decides is better for that specific study

Variable

 something that varies from one person to another

Constant

\
something that does not change

Nominal

\
* Numbers serve as tags or labels for different categories
* Numbers are not place on a meaningful scale
* higher/lower numbers don't meant anything
* Membership is both all inclusive and mutually exclusive
* A variable with only two  levels is called a **dichotomous** or **binary** variable
* ex:
* Gender (0=female) (1= male) (2=other)

Ordinal

\
* Values **are** meaningfully ordered
* Do **not** establish a numeric difference between data points
* They indicate only that one data point is ranked higher than the other 
* Distance between you 
* Ex: a student may be asked to rate a professor
* Rate from excellent to poor
* Excellent is higher than good
* Good is higher than average
* There is no fixed distance between excellent, good, or average

Interval

\
*  **Measured on a scale where each position is equidistant from the other scale points**
* Measurement intervals are equally spaced
* If the values of a variable are can be rank-ordered, and if the measurements for all the cases are expressed in the same units, then an interval level of measurement has been achieved
* Ex:
* Temperature
* 81 degrees fahrenheit is exactly 1 degree greater than 80 degrees fahrenheit

Ratio

* Interval variables with a natural zero point
* Zero simply means “none of something”
* Ex: distance
* Two objects cant be 0 inches apart
* Two objects can be 1,000 inches apart

Measurement Error

\
\-When the data we collect does not represent reality

\-always present to some degree

Random measurement errors

\
* Measurement errors that are small, non systematic, and do not threaten overall validity of the data
* Ex: a small number of survey participants misread a survey question

Systematic measurement errors

* \

**-Measurement errors are a big deal!**

**-Error in measurement in which the tool does not accurately measure the concept and is perceived incorrectly by most or all of the participants**

Reliability

\
* __Consistency__ in measurement
* Data collection instruments are reliable if they yield the same results even if used with different participants, populations, and settings
* Measurement tool should be perceived the same way by all types of participants

\

Validity

\
* The ability or potential of our data collection instrument to __actually__ measure the variable of interest
* Do our questions, tests, or other measures reflect the real meaning of the variable?

What is the difference between reliability and validity?

\
* **Reliability** pertains to measurement ability to yield consistent results
* **Validity** refers to a measurement approach ability to measure what it is supposed to

Population

\
* the group of people that you are interested in learning about

Sample

\
 *smaller portion* of a population

\
What is an inference and how is it related to why researchers use sampling?

An inference is a conclusion drawn from evidence or reasoning. Researchers use sampling to make inferences about a larger population based on a smaller sample. By studying a representative sample, researchers can make valid inferences about the population as a whole.

Probability sampling

\
* Every element of the population has a known (though not necessarily equal) chance of being selected for inclusion
* Every element has a non-zero chance of being included in the sample

Non-probability sampling

\
* **Not all elements (ex: people) of a population have an opportunity to be included in the sample**

Simple random sampling

\
* All members of a population have an equal chance of being selected for the sample
* Members of population are selected at random for inclusion in the sample

Stratified random sampling

\
* A population is divided into specific subgroups
* A random sample is subsequently drawn from each strata
* Ex: for a population with 3 strata of interest
* S1 = 5,000
* S2 = 3,000
* S3 = 2,000
* A proportional number of people selected from each strata:
* Sample S1 = 50
* Sample S2 = 20
* Sample S3 = 20

Disproportionate random sampling

Like a proportional random sampling except for the fact that sample proportions are not equivalent to population proportions

Convenience sampling

\
* Sample is drawn from those who are available or are easy to collect data from

Snowball sampling

\
* Generate a convenience sample of respondents
* Ask sampled respondents to recommend others who might be interested in providing data

Purposive sampling

\
* Researchers purposefully select from a group of people of theoretical interest
* Experts
* Extreme cases
* Typical cases

Quota sampling

\
* Generation of a sample that has attributes proportional to a given population
* Ex:
* We know that users of an internet platform are 
* 45% caucasian
* 25% asian american
* 20% african americans
* 10% other race
* Using these attributes, we can use convenience sampling techniques to construct a sample with proportional race attributes

Quantitative Content Analysis

\
* The systematic and replicable examination of symbols of communication
* The review of media materials (tv shows, magazine ads, movies, news articles, etc.) for patterns

Manifest content

\
* Content that is directly observable (not inferred or assumed)

Measurable scoring units

\
* It is important to figure out your basic or standard unit of measurement
* If you are analyzing the content of a newspaper comics page, you could take the comic strip frame as your basic unit
* If you are doing a content analysis of of magazine articles, you could take number of words
* Ex
* \
* Does a tweet contain at least one swear word?
* Does a TV commercial use sex appeal?
* Does a magazine advertisement contain a health-based appeal or claim?
* Does a newspaper article frame US foreign policy as benevolent?

What is sampled in a content analysis?

\
* Content analyses also require us to sample from a population
* The only difference to survey research is that we’re sampling media artifacts/content rather than people

Operational definition

\
* Defines how you are going to measure something

What is a “codebook” in content analysis?

a document that outlines the coding categories and definitions used to analyze data. It provides a standardized framework for researchers to systematically analyze and categorize data in a consistent and reliable manner.

What is intercoder reliability?\`

the degree of agreement or consistency between two or more coders or raters in their coding or rating of a particular phenomenon or data set. It is often used in research studies to ensure that the coding or rating process is reliable and consistent across different coders or raters.

Why is ordinal-level data problematic in content analyses?

it lacks equal intervals between categories, making it difficult to accurately measure the distance between them. This can lead to inconsistencies in data interpretation and analysis. Additionally, ordinal data does not allow for meaningful mathematical operations such as addition or multiplication, limiting the types of statistical analyses that can be performed.

What is survey research? What purpose does it serve?

a quantitative research method that involves collecting data from a sample of individuals through standardized questionnaires or interviews. The purpose is to gather information about attitudes, opinions, behaviors, and characteristics of a population. It is often used in social sciences, marketing, and public opinion research to make inferences about a larger population based on the responses of a smaller sample.

Self-report data

refers to information provided by individuals about their own thoughts, feelings, and behaviors. It is important for surveys because it allows researchers to gather information directly from the source, rather than relying on secondhand accounts or observations. It can provide valuable insights into people's attitudes, beliefs, and experiences, and can be used to inform a wide range of research questions and hypotheses.

Single selection measures

survey questions where respondents choose only one answer from a list of options. They are used to collect data on categorical variables like gender or political affiliation. They can be presented as radio buttons or drop-down menus and are used in both online and paper surveys.

Multiple selection measures

\-type of survey question that allows respondents to select more than one answer option.

\-useful when there are multiple possible answers to a question or when respondents may have more than one reason for their response.

\-It can provide more nuanced data and a better understanding of the attitudes and behaviors of the respondents.

\-Examples of multiple selection measures include checkbox questions and Likert scales with multiple options.

Ranking measure

\-survey question that asks respondents to rank items in order of preference or importance.

\-used in market research and social sciences to understand consumer preferences, opinions, and behaviors.

\-Statistical techniques can be used to analyze the results and identify patterns and trends.

Likert scale measure

\-survey rating scale that measures attitudes or opinions.

\-uses a series of statements or questions, and respondents rate their level of agreement or disagreement using a scale that ranges from strongly agree to strongly disagree.

\-responses are analyzed to provide insights into the attitudes or opinions of the respondents.

semantic differential measure

\-type of survey question that asks respondents to rate a concept or object on a series of bipolar adjectives, such as "good" vs. "bad" or "happy" vs. "sad".

\-resulting data can be used to measure attitudes or perceptions towards the concept or object being rated.

\-commonly used in social science research.

Pro of experimental research

\-detects cause and effect

Casual relationship

\-cause and effect relationship between two variables

\-one change to one variable directly affects changes to another variable

Correlation

A and B are related

Causation

A causes B

Time Order

Changes to variable A have to result in changes to variable B

**Non-Spuriousness**

the relationship between A and B must not be explained by a third variable

Independent variable

manipulated/changed by the researcher

Dependent variable

measured by the researcher

Random assignment

\- participants are randomly placed in experimental groups

\-if done to experimental groups, the groups should be similar in terms of demographic features

Experimental group

Participants who undergo a form of experimentation, such as training, taking a test or drug, or another type of intervention

Control group

Participants who do everything the experimental group members do, but are not given any test, drug, intervention, or manipulation

Randomized control trial

\-are the simplest experimental design

\-Two groups: control and experimenta

A/B test

\-A way to compare 2 versions of something to figure out which performs better

\-Considered the most basic randomized controlled experiment

Common A/B testing metrics

\-Clickthrough rate

\-Time on page

\-Bounce rate

Descriptive statistics

\-using numerical measures to summarize data characteristics, such as frequency, central tendency, and variability.

\-used to identify patterns and relationships in the data being analyzed, and to provide a summary of the data that can be easily communicated to others.

\-used to analyze and summarize text or media sources, as well as samples of numerical data.

Inferential statistics

\-using sample data to make inferences about the population of interest.

\- testing hypotheses and determining whether differences or relationships observed in the sample data are statistically significant and can be generalized to the larger population. -

\- t-tests or ANOVA are used to test hypotheses and determine the level of confidence in the results.

\-important in content analysis because they allow researchers to draw valid conclusions about the population of interest based on the sample data.