Tentafrågor

Question 1

If we wanted to understood a vaguely understood phenomen, which approach are we likely to use in our research?

Answer 1

Inductive

Question 2

The ontological perspectives positivism and constructivism (also interpretivisim, anti-posititivsm etc) can be envisioned along a spectrum, the perspectives are associated with different views on how the world works. Which of the following is more associated with a positivist perspective?

Answer 2

Measurement

Question 3

When using an inductive approach, is theory the outcome of data collection?

Answer 3

Yes

Question 4

In the Sallis et al. (2021) textbook there is a diagram showing a theoretical plane and an empirical plane. How does the operational definition of a construct relate to the theoretical definition?

Answer 4

It refines the theoretical definition into something measurable

Question 5

Does argumentation play any role in presenting research?

Answer 5

Of course. You argue for your claims based on your data, your logic, and possibly the theory you are working with.

Question 6

What are the few of the norms we follow in pursuit of conducting ethical research?

Answer 6

Not harming participants, respecting privacy, and ensuring the transparency so that the research can be evaluated.

Question 7

Ontology, epistemology, and methodology are connected. The assumptions we make regarding how we perceive reality provide guidance on how we acquire knowledge about reality, and thus, what kind of methods we apply. In the process of gaining knowledge through scientific methods, we often shift perspectives depending on the research question. From a positivt perspective, is it possible to be objective?

Answer 7

Yes

Question 8

Professors Wedlin and Sallis to some degree represent the poles of the Philosophy of Science spectrum of constructivism (Wedlin) and positivism (Sallis). Why are they both skeptical of the word “facts” and they both emphasize the importance of the word “evidence”?

Answer 8

Research, no matter which perspective, can always have flaws. In statistics, this could be represented by the trade-off between type I and type II error. Conclusions should be based on findings and evidence. Social scientists claiming facts should be treated with skepticism.

Question 9

One way to construct relevant research questions is to first consider research problems. Are we lacking knowledge about an important issue? Is there a practical problem that needs solving? Another way is to consider the cost of not solving a problem, like not solving the climate crisis. What is the role of the research question?

Answer 9

To motivate and focus the research

Question 10

Charting the territory means to learn about the topic area, which includes looking for previous conceptual papers and reserarch studies. In social sciences, research questions:

Answer 10

May motivate theory development

Question 11

When evaluating a research question, it is important that it:

Answer 11

Is relevant, researchable and represents a gap in knowledge

Question 12

Charting the territory means to learn about the topic area, which includes looking for previous conceptual papers and research studies. In social sciences, research questions:

Answer 12

May motivate theory

Question 13

Which answer best describes a case-study design?

Answer 13

It emphasizes a full contextual analysis of a few events or conditions and their interrelations

Question 14

What is a cross-sectional research design and how does it relate to causality?

Answer 14

Data is collected at a specific point in time from a cross-section of respondents, so concluding casual inference is weak

Question 15

What is the research design?

Answer 15

Is is the plan for how a research project will be conducted

Question 16

With respect to professor wedlins lecture on research design, which statement does not describe why a research design is needed?

Answer 16

Facilitates the formation of a research question that is relevant and researchable

Question 17

True experiments with randomly assigned subjects and control groups are unusual in social sciences, including economics and business studies. More often, you will see quasi- experimental designs like the Mexican music and sales of Mexican food example. The subjects (the people shopping) were not randomly assigned to shop in stores with/without Mexican music playing. In 2019 and again in 2021, the Nobel Prize winners for economics employed what is now widely known as “natural experiments”. Professor Wedlin talked about this. What is a natural experiment?

Answer 17

An event occurs to a specific group of people outside the control of the researchers, but in such a way as to resemble random assignment. Data is collected from before and after the event, and causality is established.

Question 18

Of these statistical methods, which one is associated with casual research design?

Answer 18

Regression

Question 19

You do an experiment where you photoshop a training clothes advertisement to manipulate perceived body weight. You get 100 people and randomly divide them into two groups. You show each group one of the two photoshopped advertisements. Then, they answer a short questionnaire about their perception of the advertisement and how attractive they think the model is. You use a scale where you can calculate averages.

What statistical analysis would you use to test the average difference between the groups?

Answer 19

Independent samples t-test

Question 20

You think people are fixated on weight differences of models in clothes advertising. You do an experiment where you photoshop a training clothes advertisement to manipulate perceived body weight. You get 100 people and randomly show them one of the two photoshopped advertisements. Then, they answer a short questionnaire about their perception of the advertisement and you ask their intention to buy the product (the dependent variable). What statistical analysis would you use in this experimental research design to test the effect on the dependent variable?

Answer 20

Regression analysis

Question 21

In Sweden, ice cream sales have a strong statistically significant negative correlation with umbrella sales. In Vietnam, the correlation is not significant. Why?

Answer 21

In Vietnam, people use umbrellas for both rain and shade.

Question 22

Is theory the outcome of data collection when applying an inductive approach?

Answer 22

Yes

Question 23

Research can be classified as either inductive or deductive. What sort of research methods are most commonly associated with a inductive research approach?

Answer 23

Qualitative methods

Question 24

The interview method is appropriate when:

Answer 24

The researcher wants to get deep knowledge about a certain phenomen

Question 25

Which answer would best fit the inductive approach?

Answer 25

The researcher, through a litterature review, finds that there is a lack of theory explaining a certain phenomenon.

Question 26

Which term describes combining several qualitative and/or quantitative methods?

Answer 26

Triangulation

Question 27

You do an experiment at a local COOP grocery store where you measure the sales of Mexican food for one week while playing Mexican music. Through the payment card, you track the gender of who is purchasing. You want to know whether there is a statistically significant difference between male and female purchasers for buying Mexican food. What statistical analysis or
test would you use?

Answer 27

Independent samples t-test

Question 28

You do a study amongst students in the business program about loyalty to smartphone brands. You measure their satisfaction, trust, and brand attachment to their current phone, and ask them whether they intend to buy the same brand on a scale from 1 to 7, where 1 is “absolutely no” and 7 is absolutely yes”. What statistical analysis would you use to test purchase intent?

Answer 28

Regression analysis

Question 29

The method chapter:

Answer 29

Should in detail describe how the research has been conducted, such as who did what, how, and when.

Question 30

James was surprised that people are most irritated by bees while barbequing and eating outdoors. Bees are vegetarian, wasps eat meat. To try and understand what he perceived as confusion, he developed a guide for what he wanted to discuss (based on theory), and invited 8 friends over for a one-hour discussion. What kind of data collection method is this?

Answer 30

A focus group

Question 31

When creating a questionnaire, what is a good way to make sure you properly cover the dimensions of each construct?

Answer 31

Start by looking at existing questionnaire on similar topics or theories

Question 32

Only 40 people answered a survey that was sent to 200 people, to give a response rate of 20%. The people who did not answer might be systematically different from those who did. What is this kind of bias called?

Answer 32

Non-response bias

Question 33

Assume that there are 500 people in a population and they are all on a list. You want a random sample of 50. You add the first 10 names to a hat and have a friend randomly pick one of the names. Starting with that name, you take every tenth person thereafter. In this way, you get
a sample of 50. What kind of sample is this?

Answer 33

Probability sample

Question 34

James made a survey to send to Campus Gotland students. Statistically speaking, this is
the population. He has a current list of all CG students, including their contact information. He will take a sample from the list and send out the survey. What is the list called?

Answer 34

The sampling frame

Question 35

About 30 years ago, a good friend did a survey on gay consumption habits. There were no lists of gay people, so instead, he sent his survey to 8 gay friends and asked them to pass the survey on to their gay friends. This is sort of like sending a chain letter. What kind of sample is this?

Answer 35

Snowball

Question 36

A researcher conducted interviews in order to develop a theory about customer-supplier relationships. She asked each respondent to describe the relationship in her/his own words without specific questions or prompting. What type of interview format did she use?

Answer 36

Unstructured

Question 37

For an interview guide, which format (open or structured) would best fit the inductive approach?

Answer 37

Questions should reflect the research question, open or structured is not relevant

Question 38

Validity and reliability are important in science. With respect to measurement, what are they?

Answer 38

Validity is how well a measure reflects what it intends to measure, and reliability is about the consistency of measurements.

Question 39

Which statement best describes qualitative data coding?

Answer 39

It is the process of organizing and labeling data.

Question 40

When you code your data according to the Gioia, Corely & Hamilton method, which coding approach is most appropriate?

Answer 40

You move from data-text (empirics) to higher analytical levels by aggregating and condensing.

Question 41

Coding is a creative process in which you:

Answer 41

Work with data in order to find patterns so that you can make sense of a certain phenomenon.

Question 42

One way to describe coding is finding your way from “raw data” to “making a statement”. Coding methods we looked at included Grounded Theory, Thematic Coding, and the Gioia Method. Coding is:

Answer 42

Most often inductive going from data to theory, but can be applied deductively to test a theory by looking for qualitative findings in the data.

Question 43

When drawing on existing theoretical frameworks and a priori ideas in the coding:

Answer 43

You get help with structuring your coding/analysis, but you risk “boxing in” or “shoehorning” your answers, while missing out on other patterns and explanations.

Question 44

When you are charting the territory you become, to some degree, embedded in the topic. The Gioia, Corely, and Hamilton (2012) article talked of the importance in qualitative research to, “get in there and get your hands dirty (p. 19)”. They also highlighted the risk of getting too close and “going native.” What did they mean?

Answer 44

Adopting the informant’s view and losing your higher-level perspective.

Question 45

You learned about four levels of measurement. Why do you almost always want to measure variables at the highest level possible?

Answer 45

Higher levels of measurement allow for more advanced statistics.

Question 46

In a survey about healthy living and stress, managers were asked their weight and height. Which level of measurement is this?

Answer 46

Ratio

Question 47

In a survey about managerial stress, managers were asked whether or not they have high blood pressure. Which level of measurement is this?

Answer 47

Nominal

Question 48

In a survey, managers were asked whether or not they have a management degree. Which level of measurement is this?

Answer 48

Nominal

Question 49

As the end of February approaches, the days are getting longer and warmer. What level of measurement is a temperature scale?

Answer 49

Interval

Question 50

As you write this exam, you are actually thinking about your favorite plant. It is cold in Uppsala and you are worried that it might be too cold sitting on the window sill. What level of measurement is a temperature scale?

Answer 50

Interval

Question 51

What does the Skewness and Kurtosis indicate for the variable Confidence?

Answer 51

The variable is normally distributed

Question 52

In statistics, when considering a normal distribution, what is an outlier?

Answer 52

An observation that is far out on one of the tails.

Question 53

Looking at the Frequency Table, what is the frequency of missing data for the two variables?

Answer 53

Number of missing responses / number of total participants

Question 54

In the Descriptive statistics table, what is the highest observation (value reported by a respondent) for Experience?

Answer 54

Look at column maximum -> statistic

Question 55

For Risk_Propensity, which response value in the dataset has the highest frequency?

Answer 55

Look at the most frequent and go to the left and look at the number

Question 56

This is a boxplot of the variable, number of employees in Swedish firms. What would be the best measure of the center of the data if the data is extremely abnormal distribution?

Answer 56

The median, because the data is so extremely abnormally distributed you could not rely on the mean.

Question 57

What does the median indicate for the variable company size as measured by the number of employees?

Answer 57

It is a nonparametric statistic that indicates the center of the data.

Question 58

In the Descriptives table, does the mean of .54 for gender make sense?

Answer 58

No, because it is a nominal variable

Question 59

How do correlation coefficients show causality?

Answer 59

Causality cannot be inferred from a correlation coefficient.

Question 60

You will be running a regression with the variables in the correlation matrix. Which correlations will show whether there is any danger of multicollinearity?

Answer 60

The correlations between what will be the independent variables.

Question 61

In a Pearson correlation matrix, why are the correlations on the diagonal equal to 1?

Answer 61

Because that is the value you get when you correlate a variable with itself: perfect positive correlation.

Question 62

Which type of validity refers to the degree to which two indicators of the same construct correlate?

Answer 62

Convergent

Question 63

Reliability refers to:

Answer 63

The consistency of how well something is measured.

Question 64

In regression, what is the error term?

Answer 64

Residual variance that is not explained by the regression coefficients.

Question 65

With respect to simple OLS regression, which of the following statements is correct?

Answer 65

The beta coefficient for the X variable indicates the slope of the regression line.

Question 66

In regression, if your model is too long (too many independent variables), the parameter estimates for the beta coefficients become less precise. If the model is too short, you are missing important independent variables and the parameter estimates for the beta coefficients become biased. Why is too short worse than too long?

Answer 66

Less precise estimates are random, so at least the error averages out.

Question 67

In regression, what is the residual variance?

Answer 67

It is the variance that is not explained by the regression coefficients.