199 terms

What is a questionnaire?

A formal framework consisting of a set of questions and scales designed to gather primary raw data.

What KIND OF RESEARCH mostly uses a questionnaire to collect data? Why?

Descriptive Research designs - b/c it can be turned into knowledge about a person, object or issue.

What does the researcher have to collect for predictive survey questionnaires?

The researcher has to collect a wider range of data that can be used in predicting changes in attitudes and behaviors as well as in testing hypothesis.

What is a pilot study?

a small scale version of the intended main research study, including all sub components that make up the main study, including data collection and analysis from about 50-100 representative respondents.

What is a pre-test?

is a descriptive research activity representing a small-scale investigation of 5 to 30 representative subjects, but focus is on a specific subset of the main study.

What is the difference between a pilot study and pre-test?

- A pilot study is a small scale version of the main research study, consisting of all of the parts of the main study (a lot bigger than the pre-test)

- Pre-test focuses on a specific part of the main study (a lot smaller than a pilot study)

- Pre-test focuses on a specific part of the main study (a lot smaller than a pilot study)

What are the (7) steps to a questionnaire design?

1) Confirm research objectives

2) Select appropriate data collection

3) Develop questions and scaling

4) Determine layout and evaluate questionnaire

5) Obtain initial client approval

6) Revise and finalize questionnaire

7) Implement the survey

2) Select appropriate data collection

3) Develop questions and scaling

4) Determine layout and evaluate questionnaire

5) Obtain initial client approval

6) Revise and finalize questionnaire

7) Implement the survey

What does the first step, Confirm research objectives consist of?

The research objectives are agreed upon by the researcher and client.

What does the second step, Select Appropriate Data Collection Method consist of?

1. The research must determine the data requirements/guidelines and the type of respondent demographic information

2. The researcher must follow a general to specific order.

2. The researcher must follow a general to specific order.

What are the two kinds of question formats, (in step 3) when developing questions and scaling?

1. Unstructured questions

2. Structured questions

2. Structured questions

What are unstructured questions?

Open-ended questions formatted to allow respondents to reply in their own words

- i.e. Coke reminds me of _____________________

- i.e. Coke reminds me of _____________________

What are structured questions?

Closed-ended questions that require the respondent to choose from a predetermined set of responses or scale points

- i.e. I like shopping at Costco.

(Strongly disagree, disagree, etc)

- i.e. I like shopping at Costco.

(Strongly disagree, disagree, etc)

What is important to remember about wording while developing questions and scaling?

1. Avoid ambiguous and difficult words

2. Sensitive questions should be structured carefully.

2. Sensitive questions should be structured carefully.

What are considered sensitive questions?

Include income, sexual beliefs or behaviors, medical conditions, financial difficulties, alcohol consumption, etc. that respondents are likely to respond to incorrectly.

What are considered bad questions?

1. Unanswerable

2. Leading (loaded)

3. Double-barreled

Any questions that prevent or distort the fundamental communication between the researcher and the respondents

2. Leading (loaded)

3. Double-barreled

Any questions that prevent or distort the fundamental communication between the researcher and the respondents

When would you ask sensitive questions in the questionnaire?

Later

What are skip questions?

Used if the next question or set of questions should be responded to only by respondents who meet a previous condition.

What is important to remember about forming the questionnaire (time/progress)?

Respondents should be made aware of the time taken to complete the study and their progress during the survey.

What is the general layout of the questionnaire?

1. Introductory section

2. Screening questions

3. Research question section

2. Screening questions

3. Research question section

What is the purpose of the introductory section?

- Gives the respondent an overview of the research

- Establish the legitimacy of the research (introduction of yourself/the client)

- Report the purpose of the study

- Affirm that the data will be kept secured and anonymity of the respondents identity.

- Expected time to complete.

- Eligibility criteria

- consent

- Establish the legitimacy of the research (introduction of yourself/the client)

- Report the purpose of the study

- Affirm that the data will be kept secured and anonymity of the respondents identity.

- Expected time to complete.

- Eligibility criteria

- consent

What is the purpose of the screening questions?

Identify qualified prospective respondents

- To prevent unqualified respondents form being included in the study.

- To prevent unqualified respondents form being included in the study.

When does the response order bias occur? What is it?

Occurs when the order of the questions or closed-end responses to a particular question influences the answer given.

What are the problems with online survey considerations?

1. Determining response rate

2. Recruiting participants

3. Time taken to complete the survey.

2. Recruiting participants

3. Time taken to complete the survey.

What are common online questionnaire design issues?

1. Effect of response box size on the length of participant's answer in open-ended questions.

2. Radio button versus pull down menu for responses

3. appropriate use of visuals

2. Radio button versus pull down menu for responses

3. appropriate use of visuals

How would a pretest be used on conducting a questionnaire?

1. Provides a final evaluation of the questionnaire (to understand how to revise/ how the respondents will react to the survey/ how to help the respondents react better to the survey?)

How would a pilot study be used on a questionnaire?

Small-scale version of the intended research study

How can the survey be implemented?

Self-administered or interviewer-completed.

What is a cover letter?

Separate written communication to a prospective respondent

What is the role of the cover letter?

1. Primary role - Obtain the respondent's cooperation and willingness to participate in a research project

2. To Enhance respondent willingness to complete and return a survey in a timely manner.

2. To Enhance respondent willingness to complete and return a survey in a timely manner.

What is the purpose of qualitative data analysis?

To understand the psychoanalytical or cultural phenomena using text and pictures

What is the trustworthiness of qualitative analysis based on?

The rigor of the process followed while collecting and analyzing data

What is the difference between Quantitative and Qualitative data - kind of data collected?

Qualitative: Textual and visual

Quantitative: numerical

Quantitative: numerical

What is the difference between Quantitative and Qualitative data - goals?

Qualitative: To increase understanding

Quantitative: to quantify the magnitude of variables and relationships

Quantitative: to quantify the magnitude of variables and relationships

What is the difference between Quantitative and Qualitative data - time?

Qualitative: Ongoing and frequent

What is member checking?

Asking key informants to read a researcher's report to verify that the analysis is accurate

Why does qualitative research need member checking?

Because it is inductive in nature (have to inference findings)

What is the process of collecting data in qualitative research?

1. Data is transcribed?

2. Qualitative researchers adds their interim thoughts in database, which are field notes, observations written down during the data collection effort.

3. Key respondents are asked to evaluate researcher's initial research draft.

4. This feedback is also added to the official data to be evaluated.

2. Qualitative researchers adds their interim thoughts in database, which are field notes, observations written down during the data collection effort.

3. Key respondents are asked to evaluate researcher's initial research draft.

4. This feedback is also added to the official data to be evaluated.

What is the qualitative data analysis process?

1. Data reduction

2. Data display

3. Conclusion Drawing/Verification

4. Writing the report

2. Data display

3. Conclusion Drawing/Verification

4. Writing the report

What are the interrelated processes of data reduction? (Analyzing Qualitative data: step 1) - What do you do once you have gathered the data?

Interrelated processes of data reduction are...

1. Categorization and coding of data

2. Theory building

3. Iteration and negative case analysis

1. Categorization and coding of data

2. Theory building

3. Iteration and negative case analysis

What is categorization with codes?

Categorize sections (paragraphs/ phrases/ words) of the transcript and label categorized with names and code numbers.

- Note: Some categorizations are defined before transcription (based on prior knowledge) but most come to surface during iterative process.

- Note: Some categorizations are defined before transcription (based on prior knowledge) but most come to surface during iterative process.

What are the two ways categories have to be coded?

Code sheet

Codes

Codes

What is a code sheet?

Document containing the themes or categories of a particular study

Codes?

Labels or numbers used to track categories in a qualitative study

What is a comparison?

Developing and refining theory by analyzing differences and similarities in themes, constructs or types of participants.

What does the researcher have to do when they identify a new instance of a category?

It is compared to existing categories and they decide if it belongs to existing category or if it requires a new category of its own.

What is the base that researchers use for theory building?

Is a process through which researchers build theory that is grounded or based on data

What are the two ways researchers build theories in data reduction?

1. Recursive

2. Selective coding

2. Selective coding

What is recursive?

Relationship in which a variable can both cause and be caused by the same variable.

Selective coding?

Building a story line around a core category. (Other categories are related or subsumed)

What is the induction approach?

Data -> Themes -> Categories -> Theory

What is iteration?

Working through the data several times to modify early ideas

What is memoing?

Writing down thoughts as soon as possible after each interview, focus group, or site visit

What is negative case analysis?

Deliberately looking for cases and instances that contradict the ideas and theories that researchers have been developing.

What is the role of tabulation?

- helps to quantify themes that occur repeatedly

- promotes honest research

- promotes honest research

What is important to remember about tabulation?

That its use in qualitative analyses is contraversial b/c tabulation of data may mislead readers.

Why do researchers use data displays? (Analyzing Qualitative data: step 2)

- To summarize data

- To help reduce and summarize the extensive textual data collected in the study that conveys major ideas in a compact fashion.

- To help reduce and summarize the extensive textual data collected in the study that conveys major ideas in a compact fashion.

What are the common types of data display?

1. Table that explains central themes in a study

2. Diagram that suggests relationships between variables

3. Matrix that includes quotes for various themes from representative informants

2. Diagram that suggests relationships between variables

3. Matrix that includes quotes for various themes from representative informants

What is the standard for data analysis to be considered credible?

When the results are valid and reliable

What is emic validity?

Affirms that key members within a culture or subculture agree with the findings of a research report

What is cross-researcher reliability?

Degree of similarity in the coding of the same data by different researchers.

What is credibility?

Degree of rigor, believability, and trustworthiness established by qualitative research.

What is triangulation?

- Addressing the analysis form multiple perspectives

- Using multiple data sets, researchers, and time periods

- Different kinds of relevant research informants

- Using multiple data sets, researchers, and time periods

- Different kinds of relevant research informants

What is peer review?

External qualitative methodology or topic area specialists are asked to review a research analysis.

What are the threats to drawing credible conclusions in qualitative analysis?

1. Salience of first impressions

2. Selectivity

3. Co-occurences

4. Extrapolating

5. Unreliability

2. Selectivity

3. Co-occurences

4. Extrapolating

5. Unreliability

What is the salience of first impressions?

observations of highly concrete or dramatic incidents.

Why is selectivity a threat?

leads to overconfidence, especially when trying to confirm a key finding.

Why is a co-occurrence a threat??

If a co-occurrence is taken as correlations or even as causal relationships

Extrapolating

The rate of instances in the population from those observed.

What are verbatims?

Quotes from research participants that are used in research reports.

What is a data tabulation?

counting the number of observations (Cases) that are classified into certain categories

What are the methods of data tabulation?

1. One-way tabulation

2. Cross-tabulation

2. Cross-tabulation

What is one-way tabulation?

illustrated by constructing a one-way frequency table

What is the purpose of a one-way tabulation?

1. Examines the frequency of the responses

2. Determine the amount of non response to individual questions (missing data)

3. Locate mistakes in data entry

4. Communicate the results of the research project

2. Determine the amount of non response to individual questions (missing data)

3. Locate mistakes in data entry

4. Communicate the results of the research project

When you are given an example of a frequency analysis, what do you look at to make conclusions?

Valid percent

What is the purpose of descriptive statistics?

Used to summarize and describe the data obtained from a sample of respondents

What are the measures of descriptive statistics that are used to describe data?

1. Central tendency

2. Dispersion

2. Dispersion

What is central tendency?

Finding the "center" of the numbers that are in a group of data

What is the mean?

The average

m = sum of all values/total number of values

m = sum of all values/total number of values

What is median?

The middle value

What is the mode?

The most common value

What is dispersion?

Describes how close to the mean or other measure of central tendency the rest of the values in the distribution fall

How is dispersion measured?

By the range and the variance (or standard deviation).

What is the range?

The distance between the smallest and largest values in a set of responses.

What is the variance?

The average squared deviation about the mean of a distribution of values

- uses the mean as a point of reference

- uses the mean as a point of reference

What does a small variance mean?

All the values are close to the mean

What does a large variance mean?

All the values are spread out from the mean

What is the formula to find the standard deviation?

sqrt(variance)

What is a hypothesis?

Unproven supposition or proposition that tentatively explains certain facts or phenomena.

- i.e. The average number of cokes consumed by an individual on a hot day will be greater than on a cooler day.

- i.e. The average number of cokes consumed by an individual on a hot day will be greater than on a cooler day.

What is a null hypothesis?

There is no difference in the group means. Any change is due entirely to random error.

i.e. Mean of competitive prices will not be significantly different from 5

i.e. Mean of competitive prices will not be significantly different from 5

What is an alternate hypothesis?

There is a significant difference in group means (they do not equal)

i.e. Mean of competitive prices will be significantly different from 5.

i.e. Mean of competitive prices will be significantly different from 5.

What are the steps to testing the hypothesis?

1. Always test the null hypothesis

2. Based on the analysis, we either reject the null hypothesis or fail to reject the null hypothesis

***Note: We never accept the null hypothesis.

2. Based on the analysis, we either reject the null hypothesis or fail to reject the null hypothesis

***Note: We never accept the null hypothesis.

What is a sample statistic?

Measures obtained directly from the sample or calculated from the data in the sample.

Population parameter?

Measure characteristic of the entire population.

How do you test the hypothesis?

Step 1: Develop null and alternative hypothesis.

Step 2: Select acceptable level of risk (statistical significance).

Step 3: Select appropriate statistical technique to test the hypothesis.

Step 2: Select acceptable level of risk (statistical significance).

Step 3: Select appropriate statistical technique to test the hypothesis.

How do you select the appropriate statistical technique?

Look at the factors between...

1. Number of variables to analyze

2. Scale of measurement

1. Number of variables to analyze

2. Scale of measurement

What are the (3) descriptions of the "number of variables to analyze"?

Uni variate, bi variate, multivariate

What are the three scales of measurements?

Nominal, ordinal, interval or ratio

What is a uni variate statistical test?

Analyzing one variable.

What statistical test do you use for uni variate data?

One sample t-test.

What tests do you use for the nominal scale of measurement?

Chi square

What tests do you use for the ordinal scale of measurement?

Chi-square

What tests do you use for the interval or ratio scale of measurement?

T-Test/Anova

For a uni variate statistical test, what analysis do you use?

One sample t-test

What are the steps to the one sample t-test?

1) significance of the (DONT KNOW) Slide 10 of chapter 11

How do you interpret the t-test?

1. Sig. (2-tailed) and see if it is p<0.5

2. Mean: If the mean is higher than the test value = x/midpoint of the scale.

3. If the mean is higher than the midpoint of the scale then it means the variable that you are measuring is perceived as above average.

2. Mean: If the mean is higher than the test value = x/midpoint of the scale.

3. If the mean is higher than the midpoint of the scale then it means the variable that you are measuring is perceived as above average.

What is the format to indicate that there is a significance?

t (df) = t, p= sig (2-tailed) (p-value < alpha)

What is the format to state the confidence interval?

95% confidence interval (Lower,Upper).

- 100%-alpha = confidence level

- 100%-alpha = confidence level

What is bi variate statistics?

Analyzing two variables

What happens when the p value is less than the alpha?

You reject the null hypothesis because it indicates that there is a significance.

- i.e. Respondents perceive that the food quality at this restaurant is slightly high than the midpoint of the scale. The mean 5.48 is significantly above the midpoint of 5 on a 10-point scale.

- i.e. Respondents perceive that the food quality at this restaurant is slightly high than the midpoint of the scale. The mean 5.48 is significantly above the midpoint of 5 on a 10-point scale.

What are the steps to bi variate statistical tests?

- Step 1: Null hypothesis/Alternative hypothesis

- Step 2: Significance level

- Step 3: Statistical test: Chi-square. (Look at the pearson chi-square value)

- Step 2: Significance level

- Step 3: Statistical test: Chi-square. (Look at the pearson chi-square value)

What does the chi-square measure?

Test of independence or goodness of fit

or... test for statistical significance between the frequency distributions of two (or more) nominally scaled variables in a cross-tabulation table to determine if there is any association between the variables.

or...assesses how closely the observed frequencies fit the pattern of the expected frequencies.

or... test for statistical significance between the frequency distributions of two (or more) nominally scaled variables in a cross-tabulation table to determine if there is any association between the variables.

or...assesses how closely the observed frequencies fit the pattern of the expected frequencies.

How do you determine if one variable is associated to another variable?

Look at the pearson chi-square and see if the Asymptotic significance (2-sided) is <0.05

What is the chi-square formula?

What are the degrees of freedom?

Number of independent observations upon which a quantity is based.

df = (#rows - 1) * (#columns - 1)

X1 + X2+ X3 = 50

Two variables are free and one is restricted

- Free = the number of variables in an equation or a model, that are allowed to vary.

df = (#rows - 1) * (#columns - 1)

X1 + X2+ X3 = 50

Two variables are free and one is restricted

- Free = the number of variables in an equation or a model, that are allowed to vary.

What is the difference between independent samples vs. Related samples?

The amount of samples

- Independent sample test: Two samples are given two drugs

- ex: Do coffee consumption patterns (measured by using the mean number of cups consumer daily) of males and females differ?

- Related sample test: One sample is given two drugs.

- ex: Do coffee consumption patterns differ from soft-drink consumption per day in males?

- Independent sample test: Two samples are given two drugs

- ex: Do coffee consumption patterns (measured by using the mean number of cups consumer daily) of males and females differ?

- Related sample test: One sample is given two drugs.

- ex: Do coffee consumption patterns differ from soft-drink consumption per day in males?

What is the difference between independent samples vs. Related samples?

The amount of samples

- Independent sample test: Two samples are given one drugs

- ex: Do coffee consumption patterns (measured by using the mean number of cups consumer daily) of males and females differ?

- Related sample test: One sample is given two drugs.

- ex: Do coffee consumption patterns differ from soft-drink consumption per day in males?

- Independent sample test: Two samples are given one drugs

- ex: Do coffee consumption patterns (measured by using the mean number of cups consumer daily) of males and females differ?

- Related sample test: One sample is given two drugs.

- ex: Do coffee consumption patterns differ from soft-drink consumption per day in males?

Null and alternate hypothesis for Independent samples?

- null hypothesis: no difference between the means of two groups

- Alternate hypothesis: There is a difference between the means of two groups.

- Alternate hypothesis: There is a difference between the means of two groups.

What is the equation of t (in the t-test)?

= (Difference between sample means - difference between population means) / standard error of means

What are the assumptions of the independent sample designs?

1. Data is independent in each group

2. Data is drawn from populations with normal distributions

3. Variance of the populations are equal.

2. Data is drawn from populations with normal distributions

3. Variance of the populations are equal.

How do you look at the data for an independent t-test/ independent sample?

1. Check the means of each group

2. Write down F if equal variances are assumed

3. Write down significance level: Sig <0.05, variance in two groups is not equal (assumption violated)

4. If the variances were equal, use "equal variances assumed", and use sig (2-tailed test)

2. Write down F if equal variances are assumed

3. Write down significance level: Sig <0.05, variance in two groups is not equal (assumption violated)

4. If the variances were equal, use "equal variances assumed", and use sig (2-tailed test)

When would you use the welches t-test vs. the normal t-test?

Welch's t-test: If the variance in the two groups are not equal.

Normal: If the variances are equal

Normal: If the variances are equal

How do you determine which gender is more satisfied?

See who has a greater mean.

What statistical test would you use for determining the statistical difference between three or more groups.

Anova

What is the F-ratio?

The statistical difference between three or more groups,

What is the equation of the F-Ratio?

Variance between groups/variance within groups

What is the null hypothesis for Anova (three or more groups)

There is no difference between the groups

What is the alternate hypothesis?

There is a significant difference between the groups (meaning not all means are equal)

- Note this is different than the other alternate hypothesis.

- Note this is different than the other alternate hypothesis.

Ex: Is there a difference in the satisfaction ratings among customers based on how far they have to drive to get to the restaurant?

What is the null hypothesis?

What is the null hypothesis?

There is no difference in the satisfaction ratings of customers based on how far they have to drive to get to the restaurant.

Ex: Is there a difference in the satisfaction ratings among customers based on how far they have to drive to get to the restaurant?

What is the alternate hypothesis?

What is the alternate hypothesis?

There is a significant difference in the satisfaction ratings of customers based on how far they have to drive to get to the restaurant

H1: Not all means are equal

H1: Not all means are equal

What are the steps to Anova analysis?

1. Check the means of three groups.

2. Look at F ratio

3. Look if p<0.05, shows is significant - reject the null hypothesis

2. Look at F ratio

3. Look if p<0.05, shows is significant - reject the null hypothesis

How do we formally write the F ratio?

F(df between groups, df within groups) = F

If p<0.05, and we reject the null hypothesis, what is the interpretation.

There is significant difference among the groups (customers driving from different distance)

What is the fallback from using ANOVA?

We can't tell which groups are different. All we can say is that there is a difference among the groups.

What is a paired sample t-test?

is a statistical procedure used to determine whether the mean difference between two sets of observations is zero. In a paired sample t-test, each subject or entity is measured twice, resulting in pairs of observations.

(Measure before and after)

- Suppose you are interested in evaluating the effectiveness of a company training program.

- You would measure the performance of a sample of employees before and after completing the program, and analyze the differences using a paired sample t-test.

(Measure before and after)

- Suppose you are interested in evaluating the effectiveness of a company training program.

- You would measure the performance of a sample of employees before and after completing the program, and analyze the differences using a paired sample t-test.

Ex: - Suppose you are interested in evaluating the effectiveness of a company training program.

- You would measure the performance of a sample of employees before and after completing the program, and analyze the differences using a paired sample t-test.

What is the null hypothesis?

- You would measure the performance of a sample of employees before and after completing the program, and analyze the differences using a paired sample t-test.

What is the null hypothesis?

There is no difference in the mean performance of employees before and after the training program.

Ex:

- Suppose you are interested in evaluating the effectiveness of a company training program.

- You would measure the performance of a sample of employees before and after completing the program, and analyze the differences using a paired sample t-test.

What is the alternate hypothesis?

- Suppose you are interested in evaluating the effectiveness of a company training program.

- You would measure the performance of a sample of employees before and after completing the program, and analyze the differences using a paired sample t-test.

What is the alternate hypothesis?

There is a difference in the mean performance of employees before and after the training program.

How do you analyze the Paired sample t-test?

1. Look at the means (pre and post)

2. Look at the t value --> t(df) = t

2. Look at the t value --> t(df) = t

What is the interpretation of if p <0.05?

There is a significant difference in the performance before and after the training program

What does the mean mean in the Paired samples statistics?

(Negative, reverse the sign) The average performance of the employees has increased by 6.6 units.

What is the two-way anova used for?

Two dependent variables and an interaction effect between those two independent variables.

- ex: whether customers who come to the restaurant from greater distances differ from customers who live nearby in their willingness to recommend the restaurant to a friend?

- ex: whether customers who come to the restaurant from greater distances differ from customers who live nearby in their willingness to recommend the restaurant to a friend?

Ex: Whether the willingness to recommend the restaurant is influenced by their gender?

What is the null hypothesis?

What is the null hypothesis?

There is no difference between the mean ratings of recommendations for customers who traveled different distances to come to the restaurant, and also there is no difference between males and females.

Ex: Whether the willingness to recommend the restaurant is influenced by their gender?

What is the alternate hypothesis?

What is the alternate hypothesis?

There is difference between the mean ratings of recommendations for customers who traveled different distances to come to the restaurant, and also there is no difference between males and females

How do we analyze the two way anova?

Look at the descriptive statistics:

- Split with a rectangle the distances

- Split with a circle the totals

= to understand the average to recommend the restaurant based on the distance and the gender

Ex: the average likelihood to recommend the restaurant to a friend increases as the distance increases.

Among both females and males also, the average likelihood to recommend the restaurant to a friend increases as the distance increases.

The likelihood to recommend the restaurant is higher in females than in males.

- Split with a rectangle the distances

- Split with a circle the totals

= to understand the average to recommend the restaurant based on the distance and the gender

Ex: the average likelihood to recommend the restaurant to a friend increases as the distance increases.

Among both females and males also, the average likelihood to recommend the restaurant to a friend increases as the distance increases.

The likelihood to recommend the restaurant is higher in females than in males.

What is the interaction effect?

How the variables interact with each other.

How do we search for the interaction effect?

- Look at the F-ratio for the corrected model.

Look at the p-test.

- Look at the F-ratio for the x7, gender

look at the p-test.

- look at the F-ratio for x11, distance.

look at the p-test.

Look at the p-test.

- Look at the F-ratio for the x7, gender

look at the p-test.

- look at the F-ratio for x11, distance.

look at the p-test.

How do we record the F-ratio for the corrected model?

F(df, df error) = F, p=.000

How do we record the F-ratio for the x7, gender?

F(df, df error) = F

How do we record the F-ratio for the x11, distance?

F(df, df error) = F

What is the interpretation if we fail to reject the null hypothesis?

There is no difference in males and females in willingness to recommend.

What is the interpretation if we reject the null?

There is a significant difference in the willingness to recommend between customers who travel different distances.

What is important to remember about rejecting the null and failing to reject the null hypothesis?

Have to differentiate between the variables

What are four kinds of relationships between variables?

1. Presence

2. Direction

3. Strength

4. Type

2. Direction

3. Strength

4. Type

What is presence?

If a systematic relationship exists between two or more variables.

What is direction?

Positive or negative relationship

What is strength?

SOA as no relationship, weak relationship, moderate relationship, or strong relationship.

What are the two kinds of lines that provides information about the kinds of relationships between variables?

1. Linear relationship

2. Curvilinear relationship

2. Curvilinear relationship

What is a linear relationship?

Association between two variables whereby the strength and nature of the relationship remains the same over the range of both variables

What is a curvilinear relationship?

Relationship between two variables whereby the strength and/or direction of the relationship changes over the range of both variables

What is covariance?

Amount of change in one variable that is consistently related to a change in another variable.

- Tells the degree of association between two variables

- Tells the degree of association between two variables

What are scatter plots?

Visual way to describe the relationship between two variables and the covariation they share.

What does the scatter plot look like when there is a positive linear association?

Plots in a straight line going up

What does the scatter plot look like when there is a negative linear association?

plots in a straight line going down

What does the scatter plot look like when there is nonlinear association?

plots in a u shape

What does the scatter plot look like when there is

Random plots

What is the statistical measure to analyze the correlation?

Pearson correlation coefficient

What does the pearson correlation coefficient measure?

It measures the degree of linear association between two variables.

What numbers do the pearson correlation coefficient vary between?

-1 and +1

If the pearson correlation is 0, what does this mean?

There is no relationship

What does it mean when the pearson correlation coefficient has a larger value?

Stronger association between two variables.

What is the null hypothesis for the Pearson coefficient?

The is no association between the two variables and the correlation coefficient is zero.

What if the correlation coefficient is significant?

We reject the null hypothesis

What happens if the correlation coefficient is not significant?

We fail to reject the null hypothesis

What are the three assumptions of the pearson correlation coefficient?

1. Two variables have been measured using interval or ratio scale measures (they are quantitative variables)

2. Relationship we are trying to measure is linear

3. Variables have a normally distribute population.

2. Relationship we are trying to measure is linear

3. Variables have a normally distribute population.

Ex: Relationship between satisfaction with the restaurant and likelihood to reccommend the restaurant

What is the null hypothesis?

What is the null hypothesis?

There is no relationship between satisfaction with the restaurant and likelihood to recommend the restaurant

Ex: Relationship between satisfaction with the restaurant and likelihood to reccommend the restaurant

What is the alternate hypothesis?

What is the alternate hypothesis?

There is a relationship between satisfaction with the restaurant and likelihood to recommend the restaurant.

How do we interpret the correlation coefficient from spss?

Pearson correlation/ satisfaction level

What is significant to remember while looking for the Pearson coefficient?

There is the same Pearson coefficient for both the independent variable and the dependent variable

What is the interpretation of the Pearson correlation coefficient, if the correlation coefficient is significant?

(i.e.: 601) there is a strong positive relationship between satisfaction ratings and willingness to recommend the restaurant to a friend.

What is important to remember about looking at the pearson coefficient number?

Look to see if it is positive or negative

What happens when the correlation coefficient is weak, what two possibilities must be considered??

1. There is not a consistent, systematic relationship between the two variables; or

2. The association exists, but it is not linear.

2. The association exists, but it is not linear.

What is the coefficient of determination, r^2?

Proportion of variance explained or accounted for in one variable by another.

- ex: r^2 = .3612, meaning approximately, 36.12% of the variance in willingness to recommend the restaurant is associated with satisfaction.

- ex: r^2 = .3612, meaning approximately, 36.12% of the variance in willingness to recommend the restaurant is associated with satisfaction.

What are the value ranges of the coefficient of determination?

0.00 and + 1.00

What is a bivariate regression analysis?

Statistical technique that uses information about the relationship between an independent or predictor variable and a dependent variable to make predictions.

When can we use a regression analysis?

Assuming linear relationship

Y= c + m(X) + e1

Y= c + m(X) + e1

What are the assumptions of the regression analysis?

1. Linear relationship between variables.

2. Causality is not (by default) assumed. RA defines association. Change in variable associated with another.

3. Y (response/ dependent) variable is continuous (interval or ratio).

4. Variable come from a normal population

5. Error terms are normally and independently distributed.

2. Causality is not (by default) assumed. RA defines association. Change in variable associated with another.

3. Y (response/ dependent) variable is continuous (interval or ratio).

4. Variable come from a normal population

5. Error terms are normally and independently distributed.

What does the regression analysis allow you to do?

Allows you to examine the relationship between two or more variables of interest.

How do you know to use the regression analysis?

If you need to examine the influence of one or more independent variables on a dependent variable.

- i.e. measuring the level of satisfaction and information about all the factors(independent variables)

- i.e. measuring the level of satisfaction and information about all the factors(independent variables)

What will you find in a regression analysis?

Which factors matter most, which factors can be ignored, and how these factors influence each other

What is a dependent variable?

This is the main factor that you're trying to understand or predict.

- Satisfaction from a music concert - Dependent variable

- Satisfaction from a music concert - Dependent variable

What is an independent variable?

These are the factors that you hypothesize have an impact on your dependent variable.

- Type of music, Food or catering service provided, Cost of tickets, Celebrities attending, Location, Parking, Transportation - Independent Variables

- Type of music, Food or catering service provided, Cost of tickets, Celebrities attending, Location, Parking, Transportation - Independent Variables

What does bivariate mean?

two variables

What does the error represent?

The amount of variance in satisfaction which is not explained by the ticket price.

While looking at the regression analysis, what are the steps to look at the data?

1. To see if the overall model is significant

2. To determine the regression coefficients & the Intercept.

3. To see if the regression coefficients and the intercepts are significant.

4. Form the regression equation

5. Interpret the x-intercept meaning in words.

6. Interpret the slope meaning in words.

7. Determine the r^2

8. Interpret the r^2

2. To determine the regression coefficients & the Intercept.

3. To see if the regression coefficients and the intercepts are significant.

4. Form the regression equation

5. Interpret the x-intercept meaning in words.

6. Interpret the slope meaning in words.

7. Determine the r^2

8. Interpret the r^2

In a linear regression, how do you talk about the significance of the model?

F(df regression, df residual) = F, p=x, p< or > .05

How do you say in a linear regression, that the coefficient of the regression or the intercept is significant?

1. t= ?, p = ?, p< or > .05.

When there are multiple independent variables in a regression equation, what you have to remember to include while discussing the meaning of the slope?

"Keeping all other variables constant"

How do you know if the slope is increasing or decreasing?

Look at the signs in the standardized b

How do you phrase the interpretation of the slope?

When (independent variable) increases by one unit, (dependent variable) increases/decreases by x units, keeping all other variables constant.

In a multiple linear regression, how do you interpret the intercept?

When the (independent variables) are not included in the model, the (Dependent variables) is x.