METHODOLOGY Flashcards
1
Q
(General methodology)
What is a problem statement?
A
A problem statement is the overall question you want to investigate.
2
Q
(General methodology)
What are problem questions?
A
Problem questions help you answer the problem statement. They are typically more specific than the problem statement.
3
Q
(General methodology)
What is a hypothesis?
A
An expectation of what the answer to your problem statement will be based on theory or existing knowledge. The purpose of hypothesising is to examine whether your expectation matches reality.
4
Q
(General methodology)
What is an explanatory hypothesis?
A
A reasoned claim about a relationship in reality. It consists of an observation of something in reality, such as a pattern in a table, and an explanation for the observation.
5
Q
(General methodology)
What is a unit of analysis?
A
The phenomenon or people we want to study. Units of analysis can be countries, political parties or individuals, while a unit of analysis is the individual respondent, text or interviewer.
6
Q
(General methodology)
What is value?
A
Variables can have different values.
7
Q
(General methodology)
What is operationalising?
A
Operationalising is a process of transforming abstract concepts into clearly defined variables. Operationalisation is a central part of the methodology in Social Studies. Concepts can be defined in many ways and it is important to think about how the choices you make affect your study. A good operationalisation is necessary if the study is to live up to the research criteria.
8
Q
(General methodology)
Significance
A
A significant correlation means that the correlation is not random and that it applies to the entire population.
9
Q
(Causality)
Causality
A
When one variable affects another variable. Synonym for causation
10
Q
(Causality)
Causation
A
A causation is when one variable affects another variable.
11
Q
(Causality)
Causality criteria
A
Criteria to assess whether a relation between two variables is causal.
12
Q
(Causality)
Time order
A
The independent variable must come before the dependent variable in time.
13
Q
(Causality)
Theoretical justification
A
There must be a theoretical justification for the independent variable affecting the dependent variable
14
Q
(Causality)
Empirical correlation
A
There must be an empirical correlation between the independent and dependent variable.
15
Q
(Causality)
Correlation
A
A correlation is another word for connection. In Social Studies, we use the term to describe the fact that we can see that the development of two variables follow each other. For example, if we can see that increases in social media consumption are often observed in correlation with increases in self-esteem issues. Correlation is not the same as causality.
16
Q
(Causality)
Covariation
A
Covariation means that we can see a connection between two variables. Synonymous with correlation. Covariation is not the same as causation. Two variables can co-vary without the independent variable affecting the independent variable.
17
Q
(Causality)
Spuriousness
A
We call a connection spurious if there is a covariation without causality.
18
Q
(Variable)
Variable
A
A characteristic of our units of analysis that can vary. Examples include gender, income or education.
19
Q
(Variable)
Dependent variable
A
The dependent variable is the variable we want to explain the variation in.
20
Q
(Variable)
Independent variable
A
The independent variable is the variable we believe can explain the variation in the dependent variable. The independent variable always comes before the dependent variable in time.
21
Q
(Variable)
Control variables
A
Control variables are a collective category for other types of variables we believe can affect the dependent variable.
22
Q
(Variable)
An underlying variable
A
An underlying variable is a variable that comes before both the dependent and independent variables in time and affects both the dependent and independent variables.
23
Q
(Variable)
In between variable
A
An in-between variable is a variable that comes between the independent and dependent variable in time. Affects only the dependent variable. Characterised by the fact that the effect of the independent variable goes through the in between variable. This means that the independent variable only has an indirect effect on the dependent variable. Also called an intermediate variable.
24
Q
(Variable)
Alternative independent variable
A
An alternative independent variable helps to explain the variation in the dependent variable.
25
(Variable)
Interaction variable
An interaction variable is characterised by the fact that the effect of the independent variable on the dependent variable depends on the interaction variable.
26
(The study criteria)
(The study criteria)
The study criteria are four criteria used in the social sciences to assess the quality of social science studies.
27
(The study criteria)
The four criteria
1. Validity:
Validity means that there are no systematic errors - i.e. errors that are not random - in your study.
- Internal validity: Whether the year-to-year correlation is valid.
- Measurement validity: The correlation between our theoretical concepts and the way we measure them. If there is a good correlation, validity is high.
2. Reliability: When a study has high reliability, it means that repeated measurements give the same result. Reliability means the absence of random errors, such as incorrect data entry.
3. Repeatability: A study with high repeatability means that it can be repeated so that it is possible to verify the results.
4. Generalisability: High generalisability means that we can transfer our results to something more general than what we have studied.
28
(Quantitative data)
Quantitative data
Quantitative data is information that can be counted. It can be data that has always been numerical, e.g. GDP, or data that has been transformed into numbers, e.g. by numbering the response categories in our questionnaire.
29
(Quantitative data)
Registration data
Information that has been recorded and stored in registers.
- Advantage: Easily accessible and typically population data. I.e. data about the entire population. - Disadvantage: Data may not fit exactly what we want to investigate.
30
(Quantitative data)
Survey data
Data collected via questionnaires (surveys). Usually based on a sample of the population.
- Advantage: It's easier to ask a sample of the population.
- Disadvantage: It can be difficult to ensure that your sample is representative.
31
(Quantitative data collecting)
What is a questionnaire?
A questionnaire is a survey where a representative group of people answer a series of questions.
32
(Quantitative data collecting)
What is a respondent?
A person who answers our questionnaire is a respondent.
33
(Quantitative data collecting)
What is The collective term for all the respondents relevant to a survey?
The population
34
(Quantitative data collecting)
We rarely have the opportunity to send our questionnaire to the entire population. What do we do instead?
Draw a sample. That is, an extract of the population. The larger the sample, the more representative it is. The smaller it is, the greater the statistical uncertainty.
35
(Quantitative data collecting)
What are the three types of questions used in a questionnaire?
1. Factual questions are about facts about the respondent.
2. Knowledge questions are about the respondent's knowledge and perception of the world.
3. Attitude questions are about the respondent's attitude towards a topic.
36
(Quantitative data collecting)
What is the most important thing when creating questions for a questionnaire?
The most important thing is that your questions measure your variables.
37
(Quantitative data collecting)
When creating a questionnaire, besides just creating questions, what else do you need to create?
You also need to create response categories. A response category is what the respondent can answer to a question.
38
(Quantitative data collecting)
What are the 6 different response categories?
1. Closed: If the response categories are closed, the respondent has some specific response categories to choose from.
2. Rating scale: Used to measure people's attitude towards questions or claims.
3. Ranking: Here the respondent ranks different topics.
4. Scale: The respondent is asked to place themselves on a scale, e.g. political opinion.
5. Multiple answers: The respondent must answer more than one question and provide more than one answer.
6. Open: The respondent is free to write their answer.
39
(Quantitative data collecting)
Probability-based selection
Probability-based selection means that all respondents have a known probability of being selected.
40
(Quantitative data collecting)
Simple random selection
Respondents are selected randomly by assigning everyone in the population a number and then having a computer extract the number of respondents you need.
41
(Quantitative data collecting)
Quota selection
With this selection method, respondents are selected so that the sample fulfils certain quotas. For example, quotas for gender.
42
(Quantitative data collecting)
What are the two Non-probability selections?
1. Self-selection: Respondents volunteer to answer the questionnaire.
2. Convenience selection: You select respondents based on who you have access to.
43
(Quantitative data collecting)
In Collection methods, what is a visiting interview?
The interviewer goes to the respondent's home, reads the questions aloud and notes or records the answers.
- Advantage: Respondent can ask questions. High response rate. The response rate is the proportion of the sample that completes the questionnaire.
- Disadvantage: Resource and time consuming. Risk of interviewer effects. This means that the respondent's answers are affected by the presence of the interviewer.
44
(Quantitative data collecting)
In Collection methods, what is the man on the street?
Position yourself in a busy place and ask random people as they pass by.
- Advantage: Easy to collect a large amount of data relatively quickly.
- Disadvantage: Interviewer effects. Risk that the sample is not representative.
45
(Quantitative data collecting)
In Collection methods, what is a telephone interview?
Similar to the walk-in interview, but over the phone.
- Advantage: Less time-consuming than face-to-face interviews. Respondents can ask questions.
- Disadvantage: Interviewer effects. Typically low response rate.
46
(Quantitative data collecting)
In Collection methods, what is an Internet survey?
A link is sent with the questionnaire so that respondents can complete the questionnaire online.
- Advantage: Fast and cheap.
- Disadvantage: Risk that the sample is not representative.
47
(Quantitative data collecting)
In Collection methods, what is a post-survey?
You send the questionnaire by post with a return envelope.
- Advantage: You can reach groups that are not internet savvy.
- Disadvantage: Slow and expensive method with low response rates.
48
(Quantitative data collecting)
In Collection methods, what is physical distribution?
Where you hand out the questionnaire to a group and collect it after they have responded.
-Advantage: Fast method with a high response rate.
-Disadvantage: Requires respondents to be together. Risk of them influencing each other.
49
(Quantitative data collecting)
What is a codebook?
A codebook is a tool for quantifying our data. The book is called a codebook because it contains an overview of the numerical codes we assign to each response category in our questionnaire.
50
What is Quantitative analysis?
Quantitative analysis uses numbers and statistics to investigate relationships and describe social phenomena.
51
(Quantitative analysis)
There are many different quantitative analysis methods, what is the Univariate analysis?
Univariate analysis (analysing one variable)
Median and mean are great for summarising data if you have many observations.
- Median: The median is a statistical measure of the centre value in a sorted data set.
- Average: An average is a mean value.
52
(Quantitative analysis)
There are many different quantitative analysis methods, explain the Chi2 test?
Chi2 test: A statistical measure of whether there is a relationship between our variables and whether that relationship applies to the entire population.
You interpret a chi2 test by looking at the p-value. The p-value indicates how likely it is that there is no correlation between our variables. The lower the p-value, the more likely it is that there is a correlation.
53
(Quantitative analysis)
There are many different quantitative analysis methods, explain a linear regression?
4. Linear regression: Linear regression examines whether there is a relationship between two or more variables using a straight line.
- You interpret a linear regression by looking at
The equation of the trend line: The equation of the trend line is y=ax+b, where y is the dependent variable and x is the independent variable. The A value is the interesting one when we take a reading. It shows how much the dependent variable increases or decreases when the independent variable is increased by 1.
- R2 value: This is a measure of the explanatory power of the linear regression between 0 and 1. The closer to 1, the greater the explanatory power.
- Location of points: We look at the location of the points to determine if the relationship is linear. If the points are evenly distributed around the trend line, this indicates that the relationship is linear.
54
(Quantitative analysis)
There are many different quantitative analysis methods, explain the Bivariate or multivariate analysis?
Bivariate or multivariate analysis
- Table analysis: By creating cross-tabulations, you can find relationships between your variables. Used for both bivariate and multivariate analysis.
- Crosstab: A crosstab is a table with two or more variables, where one (or more) variable is in the columns and one (or more) variable is in the rows. In the cells of the crosstab, you can see how many respondents selected each combination of response categories for our variables.
55
(Quantitative analysis)
There are many different quantitative analysis methods, explain the Bivariate analysis?
Bivariate analysis (analysing two variables)
- Proportions: A proportion is a measure of how much of a larger quantity a certain size represents.
- Index number: Index numbers tell you whether something has increased or decreased compared to the base year.
- Percentage growth: Percentage growth shows how much something has increased or decreased compared to a previous level.
56
(Quantitative analysis)
What is Statistical uncertainty?
Statistical uncertainty shows how confident we can be that the results from a sample apply to the entire population.
57
(Quantitative analysis)
What is the result of calculating statistical uncertainty?
The result of calculating statistical uncertainty is a confidence interval. This is the interval where we can be 95% certain that the result of our sample is valid for the entire population.
58
(Qualitative methodology)
What is Qualitative methodology used to?
Qualitative methodology is used to investigate qualitative data.
59
(Qualitative methodology)
What is the purpose of qualitative methodology?
The purpose of qualitative methodology is typically to understand social phenomena. You want to examine your units of analysis in-depth and get as close to the informants' experiences as possible.
60
(Qualitative methodology)
What is the strength of qualitative methodology?
The strength of qualitative methodology is that there is typically a high validity because you are trying to understand a phenomenon in depth.
61
(Qualitative methodology)
What is the weakness of qualitative methodology?
The weakness of qualitative methodology is that qualitative studies typically have low generalisability because there are typically few units of analysis.
62
(Qualitative data)
What is Qualitative data?
Qualitative data is information in text form and all human communication. Text form should be understood broadly and can be images and video as well as a speech.
63
(Qualitative data)
Qualitative data can be naturally occurring or generated, what is naturally occurring data?
Naturally occurring data is data we have not created or influenced ourselves.
- Advantage: Easily accessible data. No risk of influencing data.
- Disadvantage: Can be difficult to find data that fits our problem statement.
64
(Qualitative data)
Qualitative data can be naturally occurring or generated, what is generated data?
Generated data is data we create ourselves to answer our problem statement.
- Advantage: Data is specific to our problem statement.
- Disadvantage: Risk that we influence the data. Time-consuming.
65
(Qualitative data collecting)
Observation is a type of qualitative research that examines interaction. There are different types of observation, what is open observation?
Open observation is when you tell the people you want to study why you are there.
- Advantage: You have consent to observe.
- Disadvantage: People sometimes change their behaviour when they know they are being observed.
66
(Qualitative data collecting)
Observation is a type of qualitative research that examines interaction. There are different types of observation, what is hidden observation?
Hidden observation is when you don't tell anyone that you are there to observe.
- Advantage: The likelihood of influencing those you are observing is less.
- Disadvantage: Unethical not to ask for consent.
67
(Qualitative data collecting)
Observation is a type of qualitative research that examines interaction. There are different types of observation, what is participatory observation?
Participatory observation means that you take part in the activity you are studying.
- Advantage: Deep understanding.
- Disadvantage: Harder to be objective. Risk of bias.
68
(Qualitative data collecting)
Observation is a type of qualitative research that examines interaction. There are different types of observation, what is non-participatory observation?
Non-participatory observation means that you sit in a place where you are not in the way and observe and take notes. You don't interfere with what you are observing.
- Advantage: Easier to have an overview and stay objective.
- Disadvantage: May miss out on important knowledge.
69
(Qualitative data collecting)
An interview is a conversation between two or more people to answer a series of questions.
The people being interviewed are called informants. There are different types of interviews, what is The unstructured interview?
Here you have broad themes you want to explore, but no prepared questions.
70
(Qualitative data collecting)
An interview is a conversation between two or more people to answer a series of questions.
The people being interviewed are called informants. There are different types of interviews, what is The semi-structured interview?
You have an interview guide with themes and questions you want answers to, but you are open to the fact that during the interview it may be necessary to ask questions that are not in the interview guide.
71
(Qualitative data collecting)
An interview is a conversation between two or more people to answer a series of questions.
The people being interviewed are called informants. There are different types of interviews, what is The structured interview?
Here you follow the interview guide slavishly.
72
(Qualitative data collecting)
What are the 3 selection types to select informants?
1. Self-selection: the researcher asks informants to contact them.
2. Quota selection: The researcher has decided in advance how many informants with different characteristics are desired. For example, you want an equal number of men and women.
3. Snowball method: The snowball method involves the researcher contacting several informants who, in addition to participating in interviews, also pass on contact to more informants.
73
(Qualitative analysis)
What is content analysis?
Content analysis is used to categorise data and is done by systematically coding your data.
74
(Qualitative analysis)
What is coding data?
Coding data in qualitative methodology means that you mark words, sentences or paragraphs in the data and put a label - a code - on the piece of text. The code should be a word that describes the piece of text.
75
(Qualitative analysis)
What will data for content analysis typically be?
Data for content analysis will typically be transcribed from interviews, but can also be field notes, speeches or newspaper articles.
76
(Qualitative analysis)
A content analysis consists of three steps, which are?
Step 1: Coding the data. Coding consists of two phases:
- Open coding: Here you read the data thoroughly and apply the codes that come to mind along the way. You take an open approach to the data and code everything relevant to the problem statement. After open coding, you create a code list. The code list is a list of all the codes you have used and a short, precise definition of each code. When creating the code list, the number of codes is reduced with our theoretical concepts in mind.
- Closed coding: Here you can only use the codes from the code list.
Step 2: Analysing the data. This is done by looking for patterns and trends in the data using analytical displays. A display is a representation of qualitative data in either graphical or tabular form.
Step 3: Quality assurance and dissemination of the analysis. Here we assess our analysis against the research criteria. Typically, you will also create a dissemination display. This is a display to help the reader understand your analysis.
Bonus: You can increase the reliability of your coding through
- Intercoding: This means that you get someone else to code parts or all of the data set based on a code list.
- Intracoding: This is when you code your dataset again at a later date.
77
(Discourse analysis)
What is a Discourse analysis?
Discourse analysis is a method that can help you understand what a particular discourse consists of.
78
(Discourse analysis)
What does the term 'discourse' describe?
The term 'discourse' describes the way a certain topic is talked about in either a specific text or more broadly in a certain segment of society or a certain period in history.
79
(Discourse analysis)
What is the model for discourse analysis?
1. Find the nodal point
2. Find the chain of equivalence associated with the nodal point
3. Find the differential chain
4. Find floating point denominators
5. Consider whether the discourse is hegemonic or antagonistic
80
(Discourse analysis)
Explain the model for discourse analysis?
1. Find the nodal point
- The nodal point is the central concept within the topic that the text presents in a particular way. A text can have more than one nodal point.
2. Find the chain of equivalence associated with the nodal point
- The equivalence chain consists of the different words and phrases that the text uses to describe or define the nodal point. It is called the chain of equivalence because it consists of words or phrases that all link to the same nodal point.
3. Find the differential chain
- The difference chain is words and concepts that are contrasted with the nodal point and the equivalence chain. (This point is not always relevant.)
4. Find floating point denominators
- A floating denominator is a word or concept that is not clearly defined and is therefore open to different definitions. (This point is not always relevant.)
5. Consider whether the discourse is hegemonic or antagonistic
-A hegemonic discourse is a discourse that has become so dominant in a given society or historical context that it has become almost unthinkable not to subscribe to it.
-Antagonistic discourses are discourses that take very different approaches to the same topic.
81
(Comparative method)
What is Comparative methodology about?
Comparative methodology is about comparing units of analysis to find differences and similarities.
82
(Comparative method)
There are different ways of doing comparative methodology, what are single case studies?
You look at one case to say something general about a social phenomenon.
83
(Comparative method)
There are different ways of doing comparative methodology, what are typical case studies?
You choose a case that describes the phenomenon or causal connection you are interested in.
84
(Comparative method)
There are different ways of doing comparative methodology, what are deviant case studies?
You choose a case that does not describe the phenomenon or causal connection you are interested in.
85
(Comparative method)
There are different ways of doing comparative methodology, what is the most similar system design?
You look at many units of analysis (e.g. countries or municipalities) that are very similar but differ (vary) on the variable of interest.
86
(Comparative method)
There are different ways of doing comparative methodology, what is the most different system design?
You look at many units of analysis that are very different except for the variable we want to investigate.
