scom 388 exam 2

what are the 6 issues that researchers face when picking a research topic?

what are the 6 issues that researchers face when picking a research topic?

• empirical concern

• ethical concern

• practical concern

• is the problem significant?

• what costs and time are involved?

empirical concern

Can problems be investigated?

Ethical concern

Is there potential harm to subjects?

• What happens when we lock people in a room for weeks and make them watch violent media?

Practical concern

• Can the data be analyzed?

• Have enough practical concerns?

• Can you get access to data? Or people?

Is the problem significant?

ex: does Tv affect pets?

Can results be generalized?

Do results only apply to a subset of JMU students or students at all universities?

What costs and time are involved?

Days, years, decades?

what are the steps associated with conducting a focus group?

• define problem

• sample selection

• determine number of focus group

• study mechanics

• prepare materials

Defining problem

• Is a focus group a proper research method?

• What is already known?

• What additional info is required?

Sample selection

6 to 12 similar people participating in a focus group BUT do not know each other.

Determine number of focus groups

Typically want multiple groups with different characteristics depending on your topic.

Study mechanics

• When is it going to be held?

• Where is it going to be held?

• How are you going to record it?

• Will you offer an incentive?

• Notify participants time and place

Prepare materials

Consent form: Participants acknowledge they are participating in research, especially when recording sessions. Protocol/interview guide: A script for the moderator

what are the types of main questions in a focus group protocol

• opening question

• intro question

• transition question

• key question

• ending question

Opening question

easy, factual question that serves as an ice breaker

Intro question

General impression of the topic

• What’s your general feelings about social media? When do you use it?

Transition questions

Guide participants toward key topics

• What are the negatives and positives of using social media relative to class work?

Key questions

• Questions that address the heart of the issue

• How do you use social media to study?

• How does it hurt job prospects?

• How does it increase or decrease productivity?

Ending questions

Summarize what has been said, ask for clarification This is what you have said, correct? Anything else?

What are common problems the moderator needs to address?

Personal bias and consistency:

• Reinforce the opinions that are consistent with yours

• Look for a specific answer, so you push the conversation

pros and cons of conducting a case study

Pros: · Provides lots of detail about the case/situation. · Helps understanding of an issue. · Good at figuring out WHY something occurred in a specific situation. Cons: · Lack of generalizability · Lack of scientific objectiveness · Lack of information accessibility

what is a case study?

an up-close, in-depth, and detailed examination of a particular case

What does it mean to have a conceptual fit and provide an example of what a good conceptual fit would look like.

• You want to fit between your conceptual definition and operational definition.

• ex: a saw a good match for the job of cutting a piece of wood

In terms of measurement, what does it mean to have a valid measure?

The measure is actually measuring what you defined to measure.

In terms of measurement, what does it mean to have a reliable measure?

Measures provide consistent results.

Assessments of validity

• Face Validity

• Content Validity

• Criterion Validity

• Discriminant Validity

Face Validity

The measure seems to look good on the face of it (i.e., it looks right on the surface)

Content Validity

the measure captures the full range of meanings/dimensions of the concept (i.e., ALL aspects of the concept).

Criterion Validity

The results of your measure correspond to the results of other related measures that are considered valid

Discriminant Validity

The measures can discriminate between known different groups.

what are the ways to test reliability?

• Test – retest

• Split-hal

• Cronbach’s alpha

Test – retest

Measure the same thing twice and see if the two tests obtain the same results

Split-half

split the measures to two portions. Let some participants take one portion of the measures and others take the other portion. Then compare results between the two groups.

Cronbach’s alpha

Stat that assesses whether items get similar scores

four levels of measurement

Categorical measurement: Nominal, Ordinal

• Data are fairly simple and are language based (e.g., gender, education level) Continuous measurement: Interval, Ratio

• Data are more complex and have equivalent distance on scale (e.g., measures as inches, miles)

What were the guidelines for creating survey questions discussed in class?

• Be clear

• Keep questions short

• Avoid negatives

• Avoid double barreled questions

• Avoid leading questions

• Avoid asking questions difficult to answer

representative/probability sample

a sample from a larger group that accurately represents the characteristics of a larger population

non-representative/non-probability sample

A member of a population does not have an equal chance of being selected

Different types of non-probability samples

• Convenience

• volunteer

• Snowball

• Purposive

• Quota

Convenience

• When subjects are selected based on availability to the researcher

• Selection is generally not perfect for the purpose of research

Volunteer

• individuals volunteer to be included

• Extra credit, try new product, discounts

Snowball

• the researcher approaches one subject, ask that subject to suggest others, and it continues

• Useful for hard to reach subjects to understand social networks

Purposive

• choose certain sample for special reason

• Subjects are selected “on purpose”

• E.g., wine experts, victims of abuse

Quota

• Goal is to get the sample to look like population (e.g., 50% males, 50% females)

• Samples from the population nonrandomly

What does central tendency tell you?

The center of the data ex: average exam scores for a class

What are common measures of central tendency?

Mean, Median, and Mode

What visualization can be used to report frequency?

Pie chart, bar chart, line chart, histogram

What does dispersion tell you?

The distribution of data

• How spread out the data are

• How different are people’s responses?

What are common measures of dispersion?

Range, Deviation, Variance, Standard Deviation

What is standard deviation?

• Square root of the variance

• Most commonly used to measure how spread out the data set is

What would your data look like if you had a small standard deviation vs. a large standard deviation?

Large SD

• The values in the data set are far away from the mean, on average. In other words, the scores for people tend to be spread out. Small SD

• The values in the data set are close to the mean of the data set, on average. In other words, the scores for people to be close together.

How do inferential statistics differ from descriptive statistics?

Inferential statistics:

• Looks at two or more variables at the same time

• Allows us to make statistical inferences about the relationships between variables Descriptive statistics:

• describes data

• (e.g., mean, standard deviation) and inferential statistics allows you to make predictions (“inferences”) from that data.

What is the null hypothesis (H0)? Why do we have the null hypothesis?

• In inferential statistics, the null hypothesis is a default position that there is no relationship between two variables.

• We must start with the assumption that no relationship exists and find evidence to the contrary

• E.g., Playing violent video games does not make you more aggressive.

What is the p-value and what does it tell us?

• The p-value (or probability value): the probability of finding the null hypothesis is is true A smaller p-value means:

• Less probability that the null hypothesis is true

• Stronger evidence to support the alternative hypothesis

• If the p-value is less than .05, you are confident enough to reject the null hypothesis (and support the alternative hypothesis)

What question is correlation good at answering?

How is A related to B?

• Correlation: examines a relationship between two variables What test do we use to test for a correlation?

• The Pearson Correlation Coefficient​ (r) a.k.a Pearson’s r How do you tell the strength and direction of a correlation? / = positive; \ = negative Strength Rules of Thumb:

• Less than .3 = very weak relationship

• between .3 and .5 = weak relationship

• between .5 and .7 = moderate relationship

• over .7 = strong relationship

What question is a t-test good at answering?

• Is A greater than B?

• T-Test: tests the differences of means between two variables

What question is an ANOVA good at answering?

• What are the differences between groups? (Comparing 2+ groups)

• ANOVA = Analysis of Variance What test do we use in an ANOVA if you have more than two groups?

• post hoc test

What question is a cross-tab/chi-squared good at answering?

Is category A related to category B? Cross-tabulation

• used to test the relationship between categorical data (i.e., nominal level measure)

• A chi-square test is used to measure cross-ta

What variables do you use in a crosstab/chi-squared test?

Row variable and column variable