Exam Questions 2022-2023

Question 1

(a) State the conclusion that the Dutch book argument seeks to support. [2 points, ca. 30 words]

Answer 1

The conclusion the Dutch book argument aims to support is that rational people must have
subjective probabilities for random events that satisfy the standard axioms of probability, namely:
- Axiom 1: All probabilities are numbers between 0 and 1.
- Axiom 2: If a proposition is certainly true, it has a probability of 1. If certainly false, it has prob. 0.
- Axiom 3: If h and h* are exclusive alternatives, then P(h or h) = P(h)+P(h)

Question 2

(b) Explain how the Dutch book argument proceeds to reach its conclusion, illustrating how the argument proceeds with an example. [5 points, ca. 350 words]

Answer 2

The Dutch book argument starts from these two premises.

(P.1) If your degrees of belief do not conform to the rules of probability, there are possible betting situations where you are guaranteed to lose money, you fall prey of a Dutch book.

(P.2) You do not want to lose money. If premises P.1 and P.2 are true, then the conclusion follows that your beliefs should respect the axioms of probability.

Suppose that Your degree of belief that a toss of a coin will come out Heads is 0.6 (60%) and that Your degree of belief that the toss will come out Tails is also 0.6. In this situation, your beliefs violate the probability calculus, because, by Axiom 3, P(heads or tails) = 1.2. But Axiom 1 says that this cannot happen, since all probabilities are numbers between 0 and 1.

Now, suppose you are willing to take a bet on the outcome of the coin toss. Someone offers you

(i) a bet of €10 at 1.5:1 odds that the outcome will be heads, and
(ii) a bet of €10 at 1.5:1 odds that the outcome will be tails.

Because Your degree of belief that a toss of this coin will come out Heads is 0.6 (60%) and Your degree of belief a toss will come out Tails is 0.6 (60%), your subjectively fair odds for heads and for tails are both 1.5:1. So, you should be willing to accept both bets indeed.

But now you have accepted two bets each one of which pays worse than even money. After all, if the coin lands heads, you win €10 on the heads bet, but lose €15 on the tails bet; and the same happens if the coin lands tails. So, you are guaranteed to lose. You have been Dutch booked! So, if you don’t want to be Dutch booked and lose money, then your beliefs should not violate the axioms of probability.

Question 3

(c) Evaluate whether or not the Dutch book argument convincingly supports its conclusion.
[3 points, ca. 200 words]

Answer 3

The Dutch book argument does not provide convincing support for its conclusion in my opinion.

My opinion is supported by the following reason:

Premise P.1 is likely false for many rational agents, because betting behavior doesn’t seem to be a good guide about what anyone should believe.

First, I might dislike gambling, and so I may never gamble, which means I will never fall prey of a Dutch book.

Second, even if I liked gambling, the value I give to money
does not increase linearly, and so my betting behaviour will not readily translate into subjective probabilistic beliefs.

Third, in some cases, my belief in the outcome of some event may influence the probability of the outcome itself, which means, again, that betting behaviour about some event
will not readily translate into a probability for that event.

And finally, even assuming one satisfies the axioms of probability, that’s not a sufficient condition on rational belief, since human’s are not rational to begin with.

Question 4

(ai) State what the subjectivist interpretation of probability says. [1 point – ca. 20 words]

(aii) Explain in what sense the subjective interpretation is an epistemic interpretation of probability [2 points – ca. 40 words]

Answer 4

Subjective probability is a type of probability derived from an individual’s personal judgment or own experience about whether a specific outcome is likely to occur. It contains no formal calculations and only reflects the subject’s opinions and past experience.

Epistemic probability, is a relation between propositions: the degree to which one proposition makes another plausible. This means that all epistemic probabilities are conditional. Subjective probability is also called epistemic since there is a degree of belief in this definition is the level of confidence that the individual’s observations are at some probability to conclude a certain outcome.

Question 5

(bi) In light of an example, explain what a degree of belief is. [1 point – ca. 50 words]

(bii) In light of an example, explain how somebody’s degree of belief that a certain hypothesis is true can be revealed by possible bets that person would accept or reject. [3 points – ca. 250 words]

Answer 5

bi) A degree of belief is a way of quantifying how confident someone is in a particular statement or belief. For example, someone might say that they have a 60% degree of belief in a particular statement, meaning that they are relatively confident in its truth. This concept is often used in the field of probability and statistics to represent the uncertainty of a given event.

bii) A person’s degree of belief in a particular hypothesis can be revealed by the bets that they are willing to accept or reject. The higher the probabilities for the bet, the higher the degree of belief is the more confident the person making the bet is that they are correct.

For example, suppose that someone believes that a certain stock will go up in value. If they are willing to bet a significant amount of money on this belief, it shows that they have a high degree of belief in its truth. On the other hand, if they are only willing to bet a small amount or are unwilling to bet at all, it shows that they have a lower degree of belief in the hypothesis. This is because the amount of money that someone is willing to bet is directly tied to the confidence they have in the outcome of an event.

By examining the bets that someone is willing to accept or reject, we can get a sense of their degree of belief in a particular hypothesis.

Question 6

Explain, in light of pertinent examples, the two problems covered in our course for the subjectivist interpretation of probability. [3 points – 250 ca. words]

Answer 6

One of the problems with the subjectivist interpretation of probability is that it is highly dependent on the individual’s beliefs and subjective judgement, which can vary greatly from person to person. This can make it difficult to make objective and unbiased decisions based on probability.

Another problem with the subjectivist interpretation of probability is that it does not take into account the underlying reality or objective facts about a situation. This can lead to situations where an individual’s beliefs about the likelihood of an event are not accurate or reflective of the true probabilities of that event. For example, an individual may believe that a certain outcome is highly likely based on their own subjective judgement, but in reality, the objective probability of that outcome is much lower.

Overall, the subjectivist interpretation of probability has its limitations and can be problematic in certain situations. However, it is still a widely accepted and useful way of thinking about probability in many circumstances. This interpretation is in contrast to the more commonly held objective interpretation of probability, which sees probability as a objective property of events or outcomes.

Question 7

(ai) State what the frequentist interpretation of probability says. [1 point – ca. 20 words]

(aii) Explain why the frequentist interpretation counts as an objective interpretation of probability. [2 points – ca. 50 words]

What is the difference between frequentist and Bayesian interpretations of probability?

Answer 7

ai) The probability of an outcome is the frequency with which the outcome occurs in a long sequence of trials (maximum likelihood estimation, p values etc.). Probabilities can be found by a repeatable process of repetitions.

aii) Objective probability refers to the chances or the odds that an event will occur based on the analysis of concrete measures. Frequentist probability is a way of assigning probabilities to events that take into account how often those events actually occur. Frequentist probability is sometimes also called objective probability or empirical probability.

Frequentist statistics never uses or calculates the probability of the hypothesis, while Bayesian uses probabilities of data and probabilities of both hypothesis. Frequentist methods do not demand construction of a prior and depend on the probabilities of observed and unobserved data.

Question 8

(bi) State what the propensity interpretation of probability says. [1 point – ca. 20 words]

(bii) Explain whether or not, and why, Does the propensity interpretation counts as an epistemic interpretation of probability. [2 points – ca. 50 words]

Answer 8

bi) The propensity interpretation of probability says that the probability is thought of as a tendency of a given type of situation to yield an outcome of a certain kind, or to yield a long-run relative frequency of such an outcome.

bii) The propensity interpretation of probability is a philosophical interpretation of probability that sees probability as a measure of an inherent tendency of a given system or object to behave in a certain way. This interpretation sees probability as a property of the system or object itself, rather than as a measure of our knowledge or uncertainty about the system. As such, it can be seen as an ontological interpretation of probability, rather than an epistemic interpretation.

Question 9

(ci) Explain, in light of a pertinent example, one problem for the frequentist interpretation of probability. [2 points – ca. 200 words]

(cii) Explain, in light of a pertinent example, one problem for the propensity interpretation of probability. [2 points – ca. 200 words]

Answer 9

ci) One problem with the frequentist interpretation of probability is that it can be difficult to determine the probability of an event if the event has never occurred or if it is impossible to repeat the experiment a large number of times.

For example, consider the situation of flipping a fair coin for the first time. According to the frequentist interpretation, the probability of the coin landing on heads is determined by the relative frequency of the coin landing on heads in a large number of trials or repetitions of the experiment.

However, since the experiment has only been performed once, it is impossible to determine the probability of the coin landing on heads based on the relative frequency of that event in a large number of trials. In this case, the frequentist interpretation of probability is unable to provide any useful information about the probability of the coin landing on heads.

The frequentist interpretation of probability relies on the assumption that probability is determined by the relative frequency of an event in a large number of trials or repetitions of an experiment. However, this assumption is not always valid, as the relative frequency of an event can vary depending on the number of trials or repetitions of the experiment.

In cases where the experiment has only been performed once or where it is impossible to repeat the experiment a large number of times, the frequentist interpretation of probability is unable to provide any useful information about the probability of the event. As such, the frequentist interpretation of probability is not always the most useful or accurate way of determining the probability of an event.

cii) One problem with the propensity interpretation of probability is that it is not always clear how to determine the probability of an event based on the inherent tendencies a given system or object. For example, consider the situation of flipping a fair coin. According to the propensity interpretation, the probability of the coin landing on heads is determined by the inherent tendency of the coin to land on heads in a given situation. However, it is not clear what this inherent tendency or disposition might be, as the coin is equally likely to land on heads or tails in any given flip. In this case, the propensity interpretation of probability is unable to provide any useful information about the probability of the coin landing on heads.

This problem arises because the propensity interpretation assumes that probability is a property of a system or object, rather than a measure of our knowledge or uncertainty about the system. However, in many cases, our knowledge and beliefs about a object can provide more useful and accurate information about the probability of an event occurring than the inherent tendencies of the object itself. As such, the propensity interpretation of probability is not always the most useful or accurate way of determining the probability of an event.

Question 10

Describe the four steps we discussed in this course that are involved in Null Hypothesis Significance Testing, and illustrate each of these four steps with simple examples. [5 points, ca. 300 words]

Answer 10

1. The first step in NHST is to formulate the null hypothesis and the alternative hypothesis. For example, if a study is examining the effect of a new drug on blood pressure, the null hypothesis might be that the drug has no effect on blood pressure, while the alternative hypothesis might be that the drug has a significant effect on blood pressure.
2. Develop expectations in the form of probability distributions for possible outcomes given the truth of hypothesis. E.g., if this treatment is not effective, then when I run an experiment, such and such differences between control and treatment groups will be observed
3. Gather data/observations & Evaluate to what degree observed data violate expectations. For example how many times the drug has or does not have an effect on blood pressure.
4. Draw an inference from this comparison

Based on these observations we have a mean outcome and the significance level that determines the probability that I would obtain the observed data by “luck”. In the social sciences - significance level of .05.

Further, the p value, or probability value, tells you the probability that the null hypothesis is true.

Whether or not one should reject Null is determined by comparing the p-value and the significance level. If p-value is less than or equal to significance level, then: Reject Null hypothesis, otherwise cannot rule it out.

Null Hyp (p) leads one to expect a certain range of possible outcomes (if p, then q), when observed data are far outside that range (not-q), then we can reason such data would be very unlikely if Null is true.

Question 11

Explain how the basic logic of Null Hypothesis Significance Testing reflects the logic of the hypothetico-deductive method. [2 points, ca. 100 words]

Answer 11

In both NHST and the hypothetico-deductive method, the evidence is used to evaluate the hypothesis and determine whether it is supported by the evidence.

In NHST, this is done by calculating the probability of obtaining the observed data under the null hypothesis, known as the p-value. If the p-value is below a certain threshold (usually 0.05), the null hypothesis is rejected and it is concluded that the observed data are not due to chance and that there is a real effect or relationship present.

In the hypothetico-deductive method, the evidence is used to evaluate the scientific hypothesis and determine whether it is supported by the evidence. If the evidence supports the scientific hypothesis, it is accepted as a valid explanation for the phenomenon or observation. If the evidence does not support the scientific hypothesis, it is rejected and an alternative explanation is sought.

*The hypothetico-deductive (HD) method, sometimes called the scientific method, is a cyclic pattern of reasoning and observation used to generate and test proposed explanations puzzling observations in nature.

Identify the hypothesis to be tested.
Generate predications from the hypothesis.
Use experiments to check whether predictions are correct.
If the predictions are correct, then the hypothesis is confirmed. If not, then the hypothesis is disconfirmed.*

Question 12

Explain the two problems we discussed in our course for Null Hypothesis Significance Testing. [3 points – ca. 200 words]

Answer 12

Failure to reject the Null does not give you reason to believe that the Null is true

In an experiment, the null hypothesis and the alternative hypothesis should be carefully formulated such that one and only one of these statements is true. If the collected data supports the alternative hypothesis, then the null hypothesis can be rejected as false. However, if the data does not support the alternative hypothesis, this does not mean that the null hypothesis is true. All it means is that the null hypothesis has not been disproven—hence the term “failure to reject.” A “failure to reject” a hypothesis should not be confused with acceptance.

Choice of significance level determines the degree to which one should be willing to accept different kinds of errors

Type I error (False Positive)
Erroneously rejecting null hypothesis
Lower significance level reduces chance of type II, but raises chance of type I

Type II error (False Negative), erroneously failing to reject null hypothesis
Higher significance level reduces chance of type I, but increases chance of type II error

Significance level argued on science community and varies across the fields. In social sciences the significance level is currently 0.05.

Question 13

In 2017, a group of scientists proposed to enhance the reproducibility of the social sciences by changing the p-value threshold for statistical significance from 0.05 to 0.005.

(ai) Define what a p-value is. [1 point – ca. 25 words]
(aii) Define what reproducibility means. [2 points – ca. 80 words]

(b) Explain whether or not, and why, lowering the p-value threshold will lower the rate of Type II errors in published research. [3 points - ca. 200 words]

(c) Evaluate whether and how lowering the p-value threshold would help with the “replicability crisis.” [4 points – ca. 250 words]

Answer 13

Ai) In null hypothesis significance testing, the p-value is used as a determining factor whether a hypothesis can be rejected or accepted. In the case where you set your p-value threshold at 0.05 and the p-value that comes out of your testing comes out to 0.04, you can reject the null hypothesis with a 95% certainty.

To simplify: If your p-value is below the threshold value, you can reject the null hypothesis.

aii) Reproducibility is the ability to obtain the same results from an experiment when it is repeated using the same methods and conditions.

Important concept: it allows other researchers to verify the findings of a particular study and build upon them in future research.

Often considered to be a fundamental principle of the scientific method, as it ensures that research results are reliable and can be trusted.

b) Lowering the p-value threshold will not necessarily lower the rate of Type II errors in published research. The p-value is a measure of statistical significance, and it indicates the probability of obtaining a given result if the null hypothesis is true. A low p-value indicates that the result is unlikely to have occurred by chance,

The rate of Type II errors is influenced by a number of factors:
the sample size,
the underlying distribution of the data, and
the power of the statistical test being used.

In some cases, lowering the p-value threshold may actually increase the rate of Type II errors. This is because a lower p-value threshold makes it more difficult to reject the null hypothesis, which means that there is a greater chance of accepting the null hypothesis even when it is false. This can increased rate of Type II errors, as more false positives are accepted as true.

Therefore, while lowering the p-value threshold can help reduce the rate of Type I errors (falsely rejecting the null hypothesis), it is not necessarily effective at reducing the rate of Type II errors.

C) Lowering the p-value threshold would not necessarily help with the “replicability crisis,” which refers to the difficulty of reproducing the results of many published scientific studies. The replicability crisis is caused by a number of factors, including poor research practices, publication bias, and the use of inadequate statistical methods.

While lowering the p-value threshold can help reduce the rate of Type I errors (falsely rejecting the null hypothesis), it is not necessarily effective at improving the replicability of research results. In some cases, lowering the p-value threshold may actually make the replicability crisis worse, as it can lead to the acceptance of false positives as true results.

To address the replicability crisis, a number of changes to scientific research practices are needed. This can include:

improving the transparency and reproducibility of research methods,
increasing the use of preregistration and peer review,
promoting a culture of collaboration and openness in scientific research.

Question 14

ai) Give the formula of Bayes’s theorem. [1 point]
(aii) Explain in words what the formula of Bayes’s theorem says. [2 points – ca. 100 words]

Answer 14

ai) P(A|B) = P(B I A) * P(A) / P(B)

aii) Bayes’s theorem is a mathematical formula that describes the relationship between prior probabilities and observed evidence. It is often used to calculate the likelihood of an event based on its prior probability and new evidence.

Probability of P(A) is prior probability of the hypothesis: P(B|A) is called the posterior probability if the hypothesis. This is because P(A) is a rational degree of belief making the observation that is, prior to observation, while P(A I B) is our rational degree of belief after making the observation. the probability of event B occurring, given event A has occurred. P(A) – the probability of event A. P(B) – the probability of event B.

This allows us to update our probabilities based on new information, and to take into account the likelihood of different events given what we already know.

Question 15

(b) Describe three advantages that the Bayesian approach has over the classical approach to statistical testing, which we discussed in our course. [4 points – ca. 300 words]

Answer 15

1. Allows us to account for our previous
   knowledge of the world (priors) , since formulated competing hypotheses have been assigned a prior probability based on previous experience.
2. Allows us to check how much the data confirms or disconfirms a hypothesis. Much better than just rejecting H0

Evidence confirms H if it raises confidence that it is trueP(healthy|negative)>P(healthy)Warrants higher belief that I am healthy

Evidence disconfirms H if it lowers confidence it is trueP(healthy|positive)<P(healthy)Warrants lower belief that I am healthy

3. Informs us on how to adjust our beliefs in the different hypotheses

Question 16

(c) Explain two problems for the Bayesian approach to statistics, which we discussed in our course. [3 points – ca. 250 words]

Answer 16

There are two problems for the Bayesian approach to statistics.

The first problem is how can we define priors. This is a problem because there are often no objective criteria to define priors in hypotheses which leads to confusion. This leads to confusion because different degrees of belief (or levels of truth) turn out to be warranted (justified or guaranteed) and different priors can be assigned by different people. In other words, different subjective degrees of belief in different hypotheses can lead to different conclusions.

The second problem is that Bayesianism is not always the right approach. For some fields, abductive reasoning (making a probable conclusion from what you know) seems more apt for inquiry than Bayesian statistics. Abductive reasoning is different from deductive and inductive reasoning, which are other forms of logical inference.

Abductive reasoning is often used in scientific research to generate hypotheses and develop theories to logically arrive at a conclusion, while inductive reasoning is the process of using observations to make generalizations about a larger population. In contrast, abductive reasoning is used to arrive at the best possible explanation for a given set of observations, without necessarily arriving at a definitive conclusion.

Question 17

(a) Refer to the example of fracking and Earthquakes to describe three general characteristics, covered in our course, of causal reasoning in science and everyday life. [4 points – ca. 250 words]

Answer 17

1. cas. relationships are learned based on the timing, frequency and location of events, their correlation between these can be suggestive of a casual relationship. for example location and frequency of earthquakes
2. testing casual hypothesis involves doing something in the world, such as an intervention. leaving some factors unchanged and provide more insight into relationships and correlation of outcomes
3. cs. reasoning has great practical significance, knowing cause — how to make things and prevent things happening

Causal reasoning in science and everyday life often involves identifying and testing potential explanations for observed phenomena. In the example of fracking and earthquakes, scientists may initially observe that areas with high levels of fracking activity experience more earthquakes than expected, and may then seek to test potential explanations for this relationship, such as the idea that fracking causes the increase in earthquakes.

Causal reasoning often involves considering multiple factors that may be contributing to the observed phenomenon. In the example of fracking and earthquakes, scientists may consider not only the direct effects of fracking on the earth’s crust, but also other factors that could be contributing to the increased earthquake activity, such as natural geological processes or changes in water levels.

Causal reasoning often involves using evidence and data to support or refute potential explanations for the observed phenomenon. In the example of fracking and earthquakes, scientists may collect data on earthquake activity in areas with and without fracking, and use this data to determine whether there is a statistically significant relationship between fracking and earthquakes. This evidence can then be used to support or refute the hypothesis that fracking causes earthquakes

Question 18

(b) Describe the view that causal relationships are relationships of difference-making in
light of one example. [3 points – ca. 150 words]

Answer 18

The view that causal relationships are relationships of difference-making suggests that a cause must make a difference to the outcome or phenomenon being explained. In other words, if the cause were removed or changed, the outcome would be different.

For example, consider the relationship between smoking and lung cancer. The view that causal relationships are relationships of difference-making would suggest that smoking is a cause of lung cancer because it makes a difference to the likelihood of a person developing the disease. If a person were to stop smoking, their likelihood of developing lung cancer would be reduced. This difference in likelihood is a key aspect of the causal relationship between smoking and lung cancer.

Question 19

(c) Define the notion of a common cause, and provide one example to illustrate it. [3 points
– ca. 100 words]

Answer 19

A common cause is a third factor that can explain the relationship between two variables. It is a factor that is associated with both variables and may be the underlying reason for the observed relationship between them.

For example, consider the relationship between income and education level. It may be observed that people with higher incomes tend to have higher levels of education. However, this relationship may be explained by a common cause such as access to quality education. In this case, access to quality education is a factor that is associated with both income and education level, and may be the underlying reason for the observed relationship between these variables.

Question 20

(a) Describe the view that causal relationships are patterns regular association between
variables in light of one example. [3 points – ca. 150 words]

Answer 20

The view that causal relationships are patterns of regular association between variables suggests that when one variable consistently precedes or correlates with another variable, it may be considered a cause of that variable. For example, if researchers consistently observe that a certain medical treatment is followed by a decrease in symptoms of a particular disease, they may conclude that the treatment is a cause of the decrease in symptoms. In this case, the regular association between the treatment and the decrease in symptoms is the causal relationship.

Question 21

Explain three problems for the view that causal relationships are patterns of regular
association between variables in light of pertinent examples. [4 points – ca. 300 words]

Answer 21

One problem with the view that causal relationships are patterns of regular association between variables is that it does not take into account the potential for confounding factors. For example, if researchers observe that a certain medical treatment is consistently followed by a decrease in symptoms of a particular disease, it is possible that the treatment is not actually causing the decrease in symptoms, but rather that some other factor is responsible. For example, the patients who receive the treatment may be more likely to recover for reasons unrelated to the treatment itself.

A second problem with this view is that it may lead to incorrect conclusions about causality. For example, if researchers observe a regular association between two variables, but do not take the time to carefully control for confounding factors and conduct further analyses, they may mistakenly conclude that there is a causal relationship when in fact there is not.

A third problem with the view that causal relationships are patterns of regular association is that it does not necessarily account for the direction of causality. For example, if researchers observe a regular association between two variables, it is possible that one variable is causing the other, or that the other variable is causing the first. Without further analysis, it is impossible to determine the direction of causality from the pattern of association alone.

Question 22

State the common cause principle, and explain in what sense it is a bridge principle for causal reasoning. [3 points – ca. 100 words]

Answer 22

he common cause principle is a fundamental principle of causal reasoning that states that two events are causally related if they are both caused by a common third event. This principle is often referred to as the “bridge principle” because it allows us to bridge the gap between two events that may be correlated but do not have a direct causal relationship. For example, if we observe that a person’s symptoms improve after taking a certain medical treatment, and we also observe that the person’s symptoms tend to improve on their own over time, the common cause principle would suggest that the improvement in symptoms is due to the natural course of the disease rather than the treatment itself. By identifying a common cause for two events, the common cause principle helps us to understand the true nature of their relationship and avoid making incorrect causal inferences.

Question 23

(ai) Describe the nomological conception of explanation in light of an example. [4 points –
ca. 200 words]

Answer 23

The nomological conception of explanation is a view of explanation that focuses on the role of laws of nature in explaining the phenomena we observe in the world. According to this view, the best way to explain why something happens is to appeal to the general laws that govern the behavior of the relevant objects and forces, along with some initial conditions.

For example, suppose we want to explain why a particular apple fell from a tree. According to the nomological conception of explanation, we would appeal to the laws of gravity and motion, along with the initial conditions of the apple (such as its mass, shape, and position relative to the ground). By invoking these laws and initial conditions, we can provide a complete explanation for why the apple fell in the way that it did.

Overall, the nomological conception of explanation emphasizes the importance of understanding the general principles that govern the behavior of objects and forces in the world, and using those principles to explain specific events or phenomena. This view contrasts with other conceptions of explanation that focus on the role of cause-and-effect relationships, or the role of underlying mechanisms, in providing explanations

Question 24

Explain, in light of one example, the asymmetry problem for the nomological conception of explanation. [3 points – ca. 185 words]

Answer 24

The asymmetry problem for the nomological conception of explanation is the problem of explaining why some events are subject to explanation in terms of the laws of nature, while others are not. According to the nomological conception of explanation, the best way to explain an event is to appeal to the general laws that govern the behavior of the relevant objects and forces, along with some initial conditions. But this raises the question of why some events can be explained in this way, while others cannot.

For example, consider the phenomenon of human laughter. Why is it that we can explain why a particular person laughed at a particular joke by appealing to the laws of psychology and sociology, along with the person’s individual characteristics and the specifics of the joke? Why is it that we cannot, by contrast, explain why a rock fell to the ground in the same way?

One possible answer to this question is that human laughter is a more complex phenomenon than the falling of a rock, and therefore requires a more complex explanation that appeals to the laws of psychology and sociology. This suggests that the asymmetry problem for the nomological conception of explanation may be resolved by appealing to the complexity of the phenomena being explained.

Question 25

Explain how the causal conception of explanation solves the asymmetry problem faced by the nomological conception in light of the same example you used in part (b) [3 points – ca. 185 words].

Answer 25

The causal conception of explanation offers a solution to the asymmetry problem faced by the nomological conception of explanation, because it provides a way to explain why some events are subject to explanation in terms of the laws of nature, while others are not. According to the causal conception, the reason why some events can be explained in terms of the laws of nature is that they are caused by factors that are themselves subject to explanation in terms of those laws.

For example, consider the phenomenon of human laughter. The nomological conception of explanation would struggle to explain why we can provide a law-based explanation for human laughter, but not for the falling of a rock. But the causal conception of explanation offers a straightforward solution to this problem. We can explain why human laughter can be explained in terms of the laws of psychology and sociology by appealing to the fact that it is caused by factors (such as the content of a joke and the individual characteristics of the person laughing) that are themselves subject to explanation in those terms. In contrast, the falling of a rock is not caused by factors that are subject to explanation in terms of the laws of psychology and sociology, and therefore cannot be explained in those terms.

Question 26

(a) Define what a scientific revolution is, and give one example of a scientific revolution
discussed in our course. [2 points – ca. 150 words]

Question 27

(bi) Explain what normal science is, according to Kuhn. [2 points ca. 100 words]

(b) Explain what a paradigm is, according to Kuhn; and describe one example of a scientific paradigm covered in our course [3 points, ca. 150 words]

Answer 27

According to the philosopher of science Thomas Kuhn, normal science is the everyday work of scientists within a dominant scientific paradigm. A paradigm is a set of assumptions, values, and methods that are accepted by a community of scientists, and it provides the framework within which normal scientific research is conducted. During normal science, scientists work within the established paradigm to test hypotheses, collect data, and develop theories, focused on solving specific problems or answering specific questions.

bii) According to the philosopher of science Thomas Kuhn, a paradigm is a set of assumptions, values, and methods that are accepted by a community of scientists. A paradigm provides the framework within which normal scientific research is conducted, and it defines the problems that scientists should work on, the methods that they should use, and the standards of evidence that they should apply.

One example of a scientific paradigm is the theory of evolution by natural selection. This paradigm, which was developed by Charles Darwin in the 19th century, proposes that species evolve over time through the process of natural selection. It provides a framework for understanding the diversity of life on Earth, and it has been widely accepted by the scientific community.

The theory of evolution by natural selection is an example of a scientific paradigm because it establishes the assumptions, values, and methods that are used by scientists in a particular field of research. It defines the problems that biologists should study, the methods that they should use to study those problems, and the standards of evidence that they should apply in order to support their conclusions.

Question 28

(c) Explain Kuhn’s notion of scientific crisis in light of an example covered in our course. [3 points, ca. 150 words]

Answer 28

Kuhn’s notion of scientific crisis refers to a situation in which the dominant scientific paradigm is challenged by new evidence or ideas, and the scientific community is unable to explain or reconcile this evidence within the framework of the existing paradigm. In this situation, scientists may begin to question the fundamental assumptions and methods of the dominant paradigm, and they may develop new theories or ideas that challenge the existing paradigm.

One example of a scientific crisis occurred in the early 20th century, when the theory of relativity was developed by Albert Einstein. This theory challenged the existing dominant paradigm of classical mechanics, which was based on the idea that the laws of motion were the same for all observers. However, the theory of relativity proposed that the laws of motion were not the same for all observers, and that the speed of light was constant in all inertial frames of reference.

The development of the theory of relativity created a scientific crisis because it challenged the assumptions of the dominant paradigm of classical mechanics. Scientists were unable to explain the new evidence within the framework of classical mechanics, and many were skeptical of the theory of relativity.

Question 29

(ai) Define the value-free ideal of science. [1 point - ca. 30 words]

(aii) Explain whether the value-free ideal makes a descriptive claim about how science works or a normative claim about how science should work. [2 points - ca. 80 words]

Answer 29

The value-free ideal of science refers to the belief that scientific inquiry should be conducted without bias or preconceived notions, and that the results of scientific research should be objective and unbiased.

The value-free ideal makes a normative claim about how science should work.The value-free ideal asserts that scientific inquiry should be free from personal biases and preconceived notions, and that the results of scientific research should be objective and unbiased. This ideal is a normative claim because it provides a standard or guideline for how science should be conducted, rather than simply describing how science is actually conducted.

Question 30

(b) Explain the five questions discussed in our course that arise when doing science that values help us to answer [4 points – 300 words] (value-free ideal of science)

Answer 30

What should we study? This question may help when trying to evaluate what kind or research should be prioritized for funding. For example, members of an agency may consider a value like public health and prioritize this instead of military engineering.
How should we study it? This question may help to understand how the initial hypothesis and assumptions about the causal background both guide experimental design. For example, consider a group of researchers interested in understanding depression. There are many different ways they can study this (like the role of sleep, diet and exercise or how serotonin levels in the brain affect depression). In designing a study, researchers can just focus on one of these factors depending on their values (here understood as personal interests).
What are we trying to accomplish? This question helps getting the most accurate information versus accurate-enough information quickly enough to guide policy. For example, consider a researcher interested in learning about different aspects of students’ households in the Netherlands. There is a data set available right now, which the researcher may analyse right away; but this dataset contains inaccuracies. If the researcher waits a few months, there will be a more accurate dataset. To decide how to weigh speed vs the accuracy of the results of this study on students’ households, it is relevant to know the researcher’s values (here understood as purposes) in relation to this study.
What if we are uncertain? This question helps to understand how much evidence is required before accepting or rejecting a given hypothesis. For example, a group of agronomists decide the appropriate level for a p-value in a test concerning the safety of a new fertilizer, by asking: Given the existing situation, is the risk of a false positive (the risk of accepting that the new fertilizer works and is safe, when it is actually not safe and does not work) more serious than the risk of a false negative (the risk of rejecting the new fertilizer as unsafe and ineffective, when it actually works and is safe)?
How should we talk about the results? This question is concerned with the level of certainty conveyed to the public about some scientific finding. For example, a group of experts decide how to present the uncertainty involved in a certain hypothesis (say about the causes of poverty in the Netherlands), by asking: Given our audience of non-experts, is it more effective to present the uncertainty quantitatively as a percentage for instance, or qualitatively non-numerically and possibly with a graph?

Question 31

(c) Evaluate, on the basis of an argument, whether or not the fact that values sometimes factor legitimately into science constitute a violation of the value-free ideal. [3 points – ca. 150 words]

Answer 31

In my opinion, moral or political values that sometimes factor legitimately into science constitute a violation of the value-free ideal. Eventhough they violate this ideal, moral or political values are required for good scientific reasoning. For example, because scientific evidence may not suffice to completely prove or disprove a given hypothesis, like whether a drug is safe, scientists or public policy officers face a risk of error whenever they accept or reject that hypothesis. Scientists may wrongly accept the hypothesis as true, while the drug is not actually safe (a false positive error); or they may reject the hypothesis as false while the drug is actually safe (a false negative). To manage these two types of risks appropriately, we have to consider wheter it is better to accept the drug as safe and introduce it to the market with risk of threatening public healt or if it is better to reject the drug as unsafe with the risk of withholding a promising treatment from the public. To answer these questions it is important to consider the social value of public health. If some moral, social, and political values are sometimes relevant to set evidential thresholds for accepting/rejecting uncertain hypotheses, then some social value is sometimes required for good scientific reasoning, which shows that the value-free ideal is false also as a normative ideal.

Question 32

(ai) Define pseudoscience [1 point – ca. 20 words]

(aii) State what the falsificationist criterion says [2 point – ca. 90 words].

Answer 32

ai) Pseudoscience, which literally means “fake science,” consist of practices that deceptively appear
scientific, but are not genuinely scientific. Unlike genuine scientific practices, pseudoscience such
as, for example, scientific racism and flat earth theory, typically involves vague and ambiguous
claims, lack of coherence with other widely accepted well-confirmed scientific theories, absence of
progress and lack of openness to falsifying evidence.

aii) The falsificationist criterion is often used to distinguish between genuine scientific practices and
pseudoscience. The general idea is that scientific reasoning should proceed by attempting to
disprove ideas rather than to prove them right. More precisely, according to the falsificationist
criterion,

(i) any scientific hypothesis must be falsifiable, which means that a hypothesis is
scientific only if it is possible to say under what conditions that hypothesis will turn out to be
false, and

(ii) any scientific hypothesis/theory should be open to falsification—and when shown
false, scientists should not retain it as true.

Question 33

(b) Explain whether the falsificationist criterion, as you stated it, offers necessary or sufficient conditions (or both) for giving scientific status to a given hypothesis [3 points – ca. 300 words].

Answer 33

The falsificationist criterion provides us with a necessary, but not sufficient condition to
distinguish science from pseudoscience. This is clear from my formulation above, which says that
“a hypothesis is scientific only if it is possible to say under what conditions that hypothesis will
turn out to be false”. This formulation expresses a necessary, but insufficient cond for a hypothesis to
count as scientific. (Compare it to a requirement like “A student will pass the course only if the
student takes the exam.” Here, taking the exam is necessary but insufficient for passing.) If we
cannot in principle figure out under what conditions a given hypothesis—for example, the hyp. “Covid-19 vaccines are made out of baby elves’ organs”—would turn out to be false, then that
hypothesis cannot be falsified; and according to the falsificationist criterion, it cannot count as a
scientific hypothesis. If we can figure out under what conditions a given hypothesis—say, the hyp.
“Matteo has a degree in econometrics”—would be false, then, according to the falsificationist
criterion, we can conclude that the hypothesis is not pseudoscientific, but it may fall short of being
genuinely scientific too, if it does not comply with additional criteria like evidentialism.

Question 34

(ci) Explicitly state one hypothesis from an economic theory of your choice [1 point – ca.
50 words].

(cii) Evaluate, on the basis of an argument, whether or not the falsificationist criterion, as you stated it, would make this economic theory pseudo-scientific [3 points –200 words ca.].

Answer 34

According to the supply and demand model of the labour market described in my economics
textbook, increasing the minimum wage decreases the employment of minimum-wage workers.

One can easily figure out under what conditions it would be false that “increasing the minimum
wage of workers decreases the employment of minimum-wage workers.” This claim would be
false when increasing the minimum wage in some country does not decrease the employment of
minimum wage workers. So, because the theory I sketched above can be falsified, it respects the
falsificationist criterion, which means the supply and demand model of the labour market is a
candidate for scientific status; and so, the falsificationist criterion would offer school administrator with a good reason not to exclude it from their curricula as pseudoscientific. If the supply and demand model of the labour market also complied with additional criteria for scientificity, then school administrators would have additional good reason not to exclude it as pseudoscientific.

Question 35

(a) Define the problem of confounding and illustrate it with a simple example [2 point – ca. 100 words].

Answer 35

Confounding variables are extraneous variables that vary in an uncontrolled way and may
influence (“confound”) the relationship between the independent and dependent variable under
investigation. Example mentioned in our course: I am interested in the causal relationship
between (Ind Var) new fertilizer and (Dep Var) height of plants in my apartment. In this example,
one plausible confounding variable (Con Var) is species of plants, which, if uncontrolled, may
“confound” my conclusions about whether or not the new fertilizer causes my plants to get
higher.

Question 36

(b) Define randomization and explain how randomization differs from random sampling [2
points – ca. 80 words].

Answer 36

Randomization is one approach to indirect variable control. It consists in a chance procedure (like
flipping a coin) for assigning experimental entities (e.g., humans, animals, objects etc) to either the
treatment group or the control group.

Random sampling consists in using a chance method for selecting a sample to investigate from the
entire population. In a simple random sample, every member of the population has an equal
chance of being selected. Once you have a sample to investigate, you may use randomization to
distribute the sample between treatment and control group. Random sampling aims at creating a
sample representative of the whole population.

Question 37

(ci) Evaluate whether or not randomization is necessary for solving the problem of
confounding, supporting your position with an argument [3 points – ca. 150 words].

(cii) Evaluate whether or not randomization is sufficient for solving the problem of confounding, supporting your position with an argument [3 points – ca. 150 words].

Answer 37

Randomization helps to control for some biases (e.g., experimenter’s bias or selection bias), but it’s
strictly speaking unnecessary and insufficient for solving the problem of confounding.

ci. Randomization is unnecessary because many causally effective treatments in medicine (and in
other fields, too) were never validated in a randomized experiment – e.g., aspirin, appendectomy,
… In fact, sometimes, randomization isn’t feasible for practical or ethical reasons. If you’re
studying the effects of gestational diabetes on foetuses, for example, it’s either unethical or simply
impractical randomly assign pregnant women to an experimental condition aimed to increase
their chance of developing gestational diabetes. Randomization in this type of study is not an
option; but that does not mean the study cannot give us relevant causal evidence.

cii. Randomization is insufficient to solve the problem of confounding because random group
assignment does not guarantee that extraneous variables do in fact vary equally across the two
groups in any single experiment. It is only over an indefinite series of repetitions of the random
division that the extraneous variables will be equally distributed between the two groups. The
problem is that researchers do not make random divisions of experimental participants indefinitely
often, they do it once.

Question 38

a) Define what is a scientific model [2 point – ca. 50 words].

Answer 38

A scientific model has the aim to make a particular part of the world easier to understand, define or
visualize by referring to usually commonly accepted knowledge. It is often a simplification of a real world
scenario shaped by constraints or assumptions.

Question 39

(b) Explain the notion of idealization and the notion of abstraction [2 points – ca. 90 words].

Answer 39

The purpose of idealizations is to make the model easy to construct, manipulate, analyse, and run on a
computer. So people can focus on the important aspects of the phenomenon. Schelling wanted to make
clear that individual choices can lead (under specific conditions) to significant unintended consequences
for larger groups.

Question 40

(ci) Describe the Bay Model of the San Francisco Bay, explaining the purpose for why it was built and what one could learn about the real world from this model [3 points – ca. 200 words].

(cii) Describe one idealization and one abstraction made by the Bay Model, explaining their purpose [3 points – ca. 300 words].

Answer 40

A real life model is the model of the Bay Area, designed after an idea of John Reber. The Bay
Area model is a concrete model, which means that it is tangible. Moreover, it is a scale model, it serves as a
down-sized representation of the target system. Because the model is similar to the actual Bay Area,
scientists aim to learn more about the target system by intervening on the model as it gives a representation
of the actual Bay and portraying intervenes on the model gives a visualization to possible solutions. As
stated, scientists aim to learn more, acquiring knowledge, through a model that is similar to the real world situation. What we can learn use in real life from this model is..

i) One idealization is

One abstraction made by the model

Question 41

Consider this argument. “The choice by individual consumers to purchase or refuse to
purchase meat has no bearing on what meat producers choose to do with animals.
Therefore, anybody should be allowed to eat as much meat as they want.” This argument
contains one explicit premise and one implicit premise and is aimed at supporting one
conclusion.

(a) Reconstruct the argument in your own words, by distinguishing its two premises and its conclusion [2 points - ca. 60 words].

Question 42

Consider this argument. “The choice by individual consumers to purchase or refuse to
purchase meat has no bearing on what meat producers choose to do with animals.
Therefore, anybody should be allowed to eat as much meat as they want.” This argument
contains one explicit premise and one implicit premise and is aimed at supporting one
conclusion

(b) Given your reconstruction of the argument, explain whether or not the argument is deductively valid, providing an explicit explanation for your answer [4 points – ca. 150 words].

Question 43

Consider this argument. “The choice by individual consumers to purchase or refuse to
purchase meat has no bearing on what meat producers choose to do with animals.
Therefore, anybody should be allowed to eat as much meat as they want.” This argument
contains one explicit premise and one implicit premise and is aimed at supporting one
conclusion.

(c) Focus on one of the two premises of the argument, and describe one possible, relevant reason to believe the premise is false [4 points – ca. 200 words].

Question 44

i) Define the notion of data [1 point – ca. 50 words].

(aii) Explain in what sense “big data” are “big” [2 points – ca. 100 words]

Answer 44

ai) Data means information, more specifically facts, figures, measurements and amounts that we gather for analysis or reference in research.

aii) Big Data is used to describe the massive volume of both structured and unstructured data that is so large it is difficult to process using traditional techniques. So Big Data is just what it sounds like — a whole lot of data.

Big data can be collected from publicly shared comments on social networks and websites, voluntarily gathered from personal electronics and apps, through questionnaires, product purchases, and electronic check-ins.

Big data is most often stored in computer databases and is analyzed using software specifically designed to handle large, complex data sets.

Question 45

(b) Describe one example discussed in our course involving big data-driven research [2
points – ca. 100 words].

Answer 45

data-driven hiring systems
- to assist recruitment and review applicants
- CV’s submitted to the company over the previous 10 years
- career dynamics of company employees

What’s the problem?
> CV’s that didn’t fit the criteria for past
success were eliminated from consideration

• based on algorithms (i.e., computer-
  implementable instructions/programs)
  “trained” on “labelled” sample data to
  extract patterns or to make predictions, the company had gender-biased algorithm

> CV’s with the word “woman” were penalized

> Women ranked for less pay than men

Question 46

(c) Evaluate whether and in which sense big data-driven research always “reflects objective truth.” Support your answer by referring to pertinent examples [5 points – ca. 300 words].

Answer 46

Data are “raw material” simply “out there”
- Computational, algorithmic, statistical and mathematical techniques to analyse and gain knowledge from big data
- Has large datasets, digital format, efficiently analysable with computation tools

But
- False data
- Incomplete data
- Culturally and historically situated data

- Data should be found and stored, various legal, business, and IT constraints, including: privacy laws on data protection, commercialization of data collection and distribution, availability of suitable technology

- Should be analysed - Makes assumptions about statistical structure of data, how to weight the data and about objective of analysis - When basing it on algorithms, sample data is there to pick up patterns and make predictions - Data should be analysed carefully and does not always tell the truth

• Many assumptions on studies based on big data that might be reasonable, just like in any other study

Question 47

(ai) Define inductive inference [1 point – ca. 30 words].
(aii) Define inference to the best explanation [1 point – ca. 30 words].

Answer 47

Inductive reasoning is reasoning based on an observation, often a sample. Induction refers specifically to inference of a generalized conclusion from particular instances.

Inference may be defined as the process of drawing conclusions based on evidence and reasoning. It lies at the heart of the scientific method, for it covers the principles and methods by which we use data to learn about observable phenomena. This invariably takes place via models.

Question 48

(bi) Describe one example of inductive inference [2 point – ca. 50 words].
(bii) Describe one example of an inference to the best explanation [2 point – ca. 50 words]
(biii) Describe one example of deductive inference [2 point – ca. 50 words].

Answer 48

(Bi) For example, 25 of your 30 classmates order the same burger at a hamburger restaurant. From your
observation, you induce that the particular burger is good, otherwise, your classmates would not order it.

Bii) An inference is a conclusion that has been reached by way of evidence and reasoning. For example, if you notice someone making a disgusted face after they’ve taken a bite of their lunch, you can infer that they do not like it.

Biii) Inferences are made when a person goes beyond available evidence to form a conclusion but with a deductive inference, this conclusion always follows the stated premises. In other words, if the premises are true, then the conclusion is valid. All dogs have ears; golden retrievers are dogs, therefore they have ears.

Question 49

(c) Explain how inductive inference and inference to the best explanation differ from deductive inference [2 points – ca. 150 words]

Answer 49

The main difference between inductive and deductive reasoning is that inductive reasoning aims at developing a theory while deductive reasoning aims at testing an existing theory.

Inductive reasoning observes, seeks patterns and develops a theory. Dogs A and B have fleas - All observed dogs have fleas - All dogs have fleas. Whereas, deductive reasoning starts with an existing theory and problem statement, formulates hypothesis, collects data to test it, analyzes it and decides weather or not to reject the hypothesis

The conclusions of deductive reasoning can only be true if all the premises set in the inductive study are true and the terms are clear.A conclusion drawn on the basis of an inductive method can never be fully proven. However, it can be invalidated.

Question 50

(a) Describe the problem of induction [2 points – ca. 100 words]

Answer 50

Induction is a process of reasoning, used in science, by which a general conclusion is drawn from a set of premises, based mainly on experience or experimental evidence. .. The problem is that no inductively strong argument, no matter how strong, guarantees the truth of its conclusion. If you base your reasoning off of 9 out of 10 people saying the same thing, those 9 people could still be completely wrong. Example

Question 51

(b) Reconstruct Hume’s argument for why the general reliability of inductive inference cannot be justified [5 points – ca. 300 words]

Answer 51

Hume asks whether this evidence is actually good evidence: can we rationally justify our actual practice of coming to belief unobserved things about the world? The original problem of induction can be simply put. It concerns the support or justification of inductive methods; methods that predict or infer, in Hume’s words, that “instances of which we have had no experience resemble those of which we have had experience”

We tend to make generalizations based on inductive predictions. Hume argues that reliability of the connections since it’s possible for the conclusion to be false, even if the premises are true

Hume implies that induction must be accomplished through imagination. One does not make an inductive reference through a prior reasoning, but through an imaginative step automatically taken by the mind. We naturally reason inductively: We use experience (or evidence from the senses) to ground beliefs we have about things we haven’t observed.

We can’t really help but reason inductively. A being that was “purely rational” would never form any beliefs based upon induction, and so would never draw any generalizations or make any predictions about the future. But by nature, through the operation of custom and habit we draw inductive inferences.

In the end, Hume despairs. He sees no way to rationally justify inductive reasoning. This is a form of skepticism (about inductively acquired beliefs): We don’t have knowledge that we are tempted to think that we do. Our beliefs that come to us through inductive reasoning are in reality not rationally justifiable.

Question 52

c) Evaluate, on the basis of an argument, whether or not Hume’s argument provides us with a strong reason to believe that the general reliability of inductive inference cannot be justified [3 points – ca. 200 words].

Question 53

(a) Describe each of the six points on the checklist for evaluating whether a practice qualifies as scientific, which we discussed in our course [6 points – ca. 350 words].

Answer 53

• Naturalism — in philosophy, a theory that relates scientific method to philosophy by affirming that all beings and events in the universe (whatever their inherent character may be) are natural. Consequently, all knowledge of the universe falls within the pale of scientific investigation.
• Falsifiability, Empirical investigation — ideas that can be tested with evidence. Empirical research is defined as any study whose conclusions are exclusively derived from concrete, verifiable evidence. The term empirical basically means that it is guided by scientific experimentation and/or evidence. Likewise, a study is empirical when it uses real-world evidence in investigating its assertions
• Evidentialism — updates on ideas based on available evidence. Evidentialism is the thesis that all reasons to believe p are evidence for p. Evidentialism implies that incentives for believing p are not, thereby, reasons to believe p. For instance, evidentialism implies that the fact that believing in God would make you happy is not, thereby, a reason to believe that God exists.
• Openness to falsification — abandoned if proved wrong. The Falsification Principle, proposed by Karl Popper, is a way of demarcating science from non-science. It suggests that for a theory to be considered scientific it must be able to be tested and conceivably proven false. For example, the hypothesis that “all swans are white,” can be falsified by observing a black swan.
• Mathematical techniques — tools for use when necessary. Arithmetic, algebra and advanced mathematics may be used. Arithmetic and algebra are used to establish values and solve simple equations or formulae. In classical or everyday Physics and Chemistry, normal values are used to solve equations. In Astronomy, distances, sizes and masses are very large.Arithmetic, algebra and advanced mathematics may be used. Arithmetic and algebra are used to establish values and solve simple equations or formulae. In classical or everyday Physics and Chemistry, normal values are used to solve equations. In Astronomy, distances, sizes and masses are very large.
• Social and intitutional structure — scientific community involvement, The scientific community is key to the process of science. Members of the scientific community work together, check each other’s work, help inspire new ideas, catch cases of bias and fraud, and help motivate each other

Question 54

(b) Consider any economic theory with which you are familiar. Briefly describe the main idea of that theory and evaluate how that theory is or is not genuinely scientific based on the checklist you described [4 points – ca. 250 words].

Answer 54

A specific example of economic theory could be ‘Keynesian Economics’ which implies active government policy to manage aggregate demand in order to prevent economic recessions. The checklist points out that in order for an idea to be genuinely scientific something should provide natural explanations for natural phenomena. The government actively trying to control aggregate demand of the economy is not really a natural phenomenon. For this reason economics cannot be labeled as a genuinely scientific field.

Question 55

(ai) Define what is an experiment [1 point – ca. 30 words]
(aii) Define what is an observational study [1 point ca. 30 words]

Answer 55

a scientific test in which you perform a series of actions and carefully observe their effects in order to learn about something.

Observational studies are ones where researchers observe the effect of a risk factor, diagnostic test, treatment or other intervention without trying to change who is or isn’t exposed to it. Cohort studies and case control studies are two types of observational studies.

Question 56

(b) Define external validity and internal validity and describe the importance of each [3 points – ca. 200 words]

Answer 56

Internal validity examines whether the study design, conduct, and analysis answer the research questions without bias. External validity examines whether the study findings can be generalized to other contexts

There are two main types of external validity: population validity and ecological validity. Population validity refers to whether you can reasonably generalize the findings from your sample to a larger group of people (the population). Ecological validity refers to whether you can reasonably generalize the findings of a study to other situations and settings in the ‘real world’. Without high external validity, you cannot apply results from the laboratory to other people or the real world.

Internal validity makes the conclusions of a causal relationship credible and trustworthy. Without high internal validity, an experiment cannot demonstrate a causal link between two variables. You need to be able to rule out other explanations (including control, extraneous, and confounding variables) for the results. Without high internal validity, an experiment cannot demonstrate a causal link between two variables.

Question 57

(c) Evaluate the main advantages and disadvantages of a laboratory experiment compared to a field experiment [5 points – ca. 300 words]

Answer 57

You want to test the hypothesis that driving reaction times become slower when people pay attention to others talking. In a laboratory setting, you set up a simple computer-based task to measure reaction times. In In the example above, it is difficult to generalize the findings to real-life driving conditions. A computer-based task using a mouse does not resemble real-life driving conditions with a steering wheel.

To improve ecological validity in a lab setting, you could use an immersive driving simulator with a steering wheel and foot pedal instead of a computer and mouse. This increases psychological realism by more closely mirroring the experience of driving in the real world.

Alternatively, for higher ecological validity, you could conduct the experiment using a real driving course.

Advantages
Greater ecological validity than laboratory experiment.
Less sample bias.
Fewer demand characteristics if participants are unaware.

Disadvantages
Lack of control brings problem of extraneous variables.
Difficult to replicate.
Difficult to record data accurately.
Ethical problems.