L1,L2,L3,L5,L6,L9 Flashcards

Question

What is a cause?

Answer 1

Most causes are more accurately called inus conditions. Presence of a match → forest fire? Any given effect often requires many factors to occur, but we rarely know all of them, or how they relate to each other. As a consequence, many causal relationships: are probabilistic rather than deterministic in a strict sense are not universal, but occur only under some conditions or in certain contexts

Answer 2

It is rare that causal relationships hold under all conditions with all types of people and at all historical time periods. It is vital to identify relationships that hold more consistently.

Answer 3

An effect can be understood as the difference between what actually happened (the fact) and what would have happened if a specific cause had not occurred (the counterfactual). Effect: The effect is essentially the difference between what actually happened (fact) and what would have happened if the cause had not occurred (counterfactual).

Answer 4

The cause preceded the effect The cause was related to the effect We can find no plausible alternative explanation for the effect, other than the cause In experiments we: Manipulate the presumed cause (and observe the outcome) Observe whether variation in the cause is related to variation in the effect Attempt to rule out other explanations of the effect

Answer 5

We may not know which variable came first (i.e., whether the presumed cause preceded the effect). We cannot rule out alternative explanations. If there exists a third variable that causes the effect, this is called a confounding variable, or simply confound.

Answer 6

Cause This is the variable that is manipulated by the experimenter The variable that i change We want to know is if the independent variable affects the dependent variable in some way.

Answer 7

Effect (result of experiment) This is the variable that is measured by the experiment It depends on the independent variable The dependent variable is the variable that is observed and measured to see if it is affected by changes in the independent variable.

Answer 8

A hypothesis is a precise problem statement that can be directly tested through an empirical investigation. Example: “The iOS virtual keyboard is faster and more accurate than the Android virtual keyboard.”

Answer 9

A true experiment must have randomization. Random assignment is intended to create units that are probabilistically similar to each other on average Example: Every individual has the same chance to end up in a given condition. Therefore, any observed differences in outcome are likely to be due to the treatment (= cause)

Answer 10

Simulation refers to the process of creating and analyzing a model or imitation of a real-world system or process to gain insights into its behaviour or performance. It involves using a computer program or mathematical model to replicate a complex system's essential features and dynamics, allowing researchers to observe, experiment, and make predictions in a controlled and virtual environment.

Answer 11

Modeling: Constructing a representation of the real-world system or phenomenon using mathematical equations, algorithms, or computer programming. Manipulation: Changing variables or parameters in the simulated environment to observe their effects on the system's behavior. Analysis: Collecting and interpreting data generated by the simulation to draw conclusions about the real-world system. Prediction: Extrapolating the simulation results to make predictions about the future behavior or outcomes of the real-world system under different conditions.

Answer 12

A variable that is not controlled but can affect the dependent variable.

Answer 13

A control variable is a circumstance (not under investigation) that is kept constant while testing the effect of an independent variable More control means the experiment is less generalizable (i.e., less applicable to other people and other situations)

Answer 14

Each participant is tested on each condition Within-subjects advantages: Fewer participants Less “variation due to participants” No need to balance groups (because there is only one group!)

Answer 15

Each participant is tested on one condition only Between-subjects disadvantage More participants (harder to recruit, schedule, etc.) More “variation due to participants” Need to balance groups (to ensure they are more or less the same)

Answer 16

Internal validity threats * Reasons why inferences (assumption) that the relationship between two variables is causal may be incorrect External validity * Reasons why inferences about how study results would hold over variations in persons, settings, treatments, and outcomes might be incorrect.

Answer 17

Any means of asking people for information

Answer 18

With a paper or web-based survey, participants answer the questions on their own - No chance to clarify or ask follow-ups - It is hard to know how questions are interpreted - Once the survey is out it is out

Answer 19

We use surveys when we need to investigate a clearly defined topic that involves measuring - Knowledge - Opinions - Behaviours - Perception

Answer 20

We need our surveys to have a few important properties - Relevant to our research questions - Easy to interpret - Be interpreted in the same way by all participants general steps to survey development - Define variables to measures - Develop/adapt/find appropriate questions to measure those - Validate and pilot

Answer 21

The first step in developing a survey is to define variables. Then, come up with questions to measure those variables. Often we use several questions for the same variable, to capture different aspects of the variable topic These questions can be: - Knowledge questions - Likert scale questions When developing questions there are some things to consider - Clarity - Consistency - Question order - Response order

Answer 22

Validity concerns how well the survey measures what it intends to measure Expert review is one typical approach - Focus on validity - Ask domain experts about the survey questions (supervisor, peers) Reliability concerns how consistently it does that Distribute the survey to a small set of respondents and test how it goes - Typically adding a comment question - Focus on reliability (are people answering as expected

Answer 23

Our target group is called population - and we can typically not ask everyone in that population. Instead, we need to draw a sample. A sample is a subset of our population which we assume to be representative of the population at large.

Answer 24

Random sample - We sample at random, and give everyone an equal chance of being selected - Practically, we could select names at random from a list of everyone in the population - Gold standard – but expensive and often infeasible

Answer 25

Stratified sampling (a practical approach to random sampling) - The population is split into sub-population based on criteria such as age, gender…. - A random sample is drawn from each sub-population - Intends to ensure a greater spread over the population

Answer 26

Convenience sampling - Aka accidental sampling or opportunity sampling - Involves recruiting participants “close at hand” - Includes asking your friends, using social media, etc - We cannot know how representative the sample is of the target population

Answer 27

Purposive sampling - Selective sampling where “high-value” respondents are invited to answer - The idea is to include participants who are likely to generate valuable data - Useful in some cases, but hard to generalize from the results

Answer 28

Sampling bias is skewness in results due to improper sampling. Confirmation bias - Participants tend to agree rather than disagree with opinions - Try to avoid opinionated questions Question order bias - Question order may impact how participants answer - Randomize questions to avoid Survey fatigue bias - Participants may get fatigued and stop answering properly - Limit the survey length

Answer 29

Clear: - Easy to understand, impossible to misunderstand Concise: - Anything not needed for the task at hand should be omitted Objective: - Reflection of the work

Answer 30

Academic writing has a purpose (contribute to scientific knowledge), specific audience (scientists, academics), tone (unemotional, unbias)

Answer 31

Introduction - Introduce the relevant literature - Problem description Materials and methods - Introduce the study system - What research methods have been used? - How was the study done? Results - Objectively state finds - Focus on results Discussion - Interpret you result - Tie result back to the literature by answering the knowledge gap Conclusions and implications

Answer 32

Bias comes from you and your design, and it can affect the results based on several issues: e.g. chosen sample, investigation method and style of analysis.