Question 3 is the question that involves a higher evel of thinking.  Notice that all of the command terms used so far are rather "low level" command terms.  Question 3 asks you to discuss one question with regard to the stimulus piece. The question will be one of the following:

Discuss the possibility of generalizing the findings of the study.

Discuss how a researcher could ensure that the results of the study are credible.

Discuss how the researcher in the study could avoid bias.

Question three is assessed using the following assessment rubric.

MY REFLECTION-THE MARKED DIFFERENCE BETWEEN 4-9 and 7-9 is the specific link to the stimulus with its proper use through words, phrases, and quotes from the stimulus and the framing of the argument with the use of transition connectors. To move to 7-9, I, as a student, would have to frame the argument more strong and use the stimulus very effectively in drawing references while writing appropriate knowledge.

Question 1b) 2+1= (2 for identifying and explaining the sampling technique and 1 for linking to the stimulus)

Question 1c) 3 is for coming up with a fair research method (1)for the stimulus context and justifying the method (2)

Question 1 is actually not, in fact, one question - but three.  You are always asked the same three questions.  These questions are:

Identify the method used and outline two characteristics of the method.

Describe the sampling method used in the study.

Suggest an alternative or additional research method giving one reason for your choice.

Each question is worth three marks.

Question 2

For question 2 the focus is ethics. You will be asked one of the following questions:

Describe the ethical considerations that were applied in the study and explain if further ethical considerations could be applied.

Describe the ethical considerations in reporting the results and explain ethical considerations that could be taken into account when applying the findings of the study.

The question is worth six marks. Up to [2 marks] awarded for describing the ethical considerations identified in the stimulus piece: anonymity and debriefing. Up to [4 marks] awarded for describing relevant ethical considerations that are not mentioned in the stimulus piece. If the further ethical considerations are simply described and not explained, then a maximum of [2 marks] should be awarded.

Question 1

A. Identify the method used and outline two characteristics of the method.

The method used in this study was a case study. A case study often focuses on a single group - in this a single school - over a period of time in order to gather a large amount of rich data. In addition, case studies use method triangulation - that is, more than one research method is used as part of the investigation.  In this study, both focus groups and observations were used.

B. Describe the sampling method used in the study.

The sampling method used was a voluntary or self-selected sample. This was done most likely to find a school who would be willing to have the disruption of the researchers during the school day. The sample may have also been purposive in that the researchers may have specified in the ad how big the school should be, how long the IB program has existed at the school, and whether the school is public or private.

C. Suggest an alternative or additional research method giving one reason for your choice.

The researchers could also carry out an experiment with the students to see if their perceptions of their own stress are accurate. The experiment could be a natural experiment around mock exam time.  The researchers could measure the students' level of cortisol and white blood cell count.  This would measure both the level of stress and its effect on their immune system.  The measurements would be taken both before and after the mock exams. This additional method would not establish a cause-and-effect relationship, but it could confirm the self-reports of the students' level of stress.

Question 2

Describe the ethical considerations that were applied in the study and explain if further ethical considerations could be applied.

The researchers had to obtain consent in order to carry out the study.  Parental consent was obtained, perhaps in line with the school's policy. As part of the consent, the researchers would have to explain the aim of their research and how they would gather their data as well as explain to the participants their rights.  In addition, the researchers debriefed the participants.  This would be a chance to explain what they had learned from their research, give the participants the right to withdraw their data, and remind them of their rights.  In the study, the researchers also provided information about coping with stress in order to assist those that experienced high levels of anxiety.

In addition, the researchers would have kept all information anonymous.  This includes not only the participants' comments in the focus groups and interviews but also the individual students' anxiety ratings.  It is also important that any descriptions of students do not make it possible for them to be identified. Another concern is the audio recording of the focus groups.  These recordings should not be used in a public forum which might allow the participants to be recognized.

Question 3

Discuss the possibility of generalizing/transferring the findings of the study.

Often a case study has the goal of helping the population that is being studied.  One goal of the researcher is then representational generalizability. This means that the students chosen for this study would be representative of the population that they were taken from and the findings made by the researchers can be then generalized to the larger school population. In this case, we may be able to generalize the findings to other students in the program.  The sample was quite diverse, so this may help to generalize the findings, but it is also not clear from the study how these particular students were chosen. If they were all taken from the same class - e.g. HL English - then they may not be representative of the population and so representational generalizability would be low.

When discussing the generalizability of a case study, psychologists also refer to its inferential generalizability or transferability. This is when the findings can be applied to another setting. To do this, the researchers would have to determine which variables would be important in the context of this study.  For example, it would have to be another IB school, with a similar population (socioeconomic, culturally) and would have to have a similar IB program.

Finally, a case study can be generalized to existing theory - that is, theoretical generalizability. This is when a theory is generated from the study which is then supported by carrying out other case studies and getting similar results.  If the researchers found that students in other IB schools had the same type of coping strategies, then the findings could be more credibly generalized.

There are several different sampling techniques. Here are the most commonly used techniques.

Self-selected sampling

Also known as a volunteer sample. People that sign-up are usually motivated and are less likely to drop out of the research. However, that may lead to a sample that is not representative of the target population.

Opportunity sampling

Also known as a convenience sample. This is when you use a pre-existing sample - for example, Mr. Smith's grade 9 English class. It is an easy way to get participants because they are readily available. The groupings often tend to be relatively homogeneous. However, they are also not usually representative of a target population. However, if you were doing a study only on grade 9 students at your school, and you used four out of the five existing English classes, then your study would be fairly representative.

Random sampling

A random sample is defined as a sample in which everyone in the target population had the same probability of being chosen. This is often seen as an ideal sampling technique, but if the target population is too large, then random sampling is not really possible. In addition, if the target population is very heterogeneous, then random sampling could still generate a biased sample.

If I were taking a random sample from my psychology students, I would put all of their names into a hat. When I withdrew a name, I would put the name back into the hat to make sure that everyone had the same probability of being chosen. Another method is to use a random number generator to create a sample.

Please note that there is a difference between random sampling vs. random allocation. Random allocation is how I decide which group or condition a participant will be in. This can also be done with a random number generator or the flip of a coin.

Purposive sampling

A purposive sample is looking for people with a very specific set of traits. This could be African-American women between 30 and 40 years old or expatriates who have lived in Prague for at least 25 years. This is often accomplished by a self-selected sample.

Snowball sampling

Also known as a network sample. This is another type of purposive sampling, but network sampling is used when you are looking for participants from a specific group that would not respond to an ad in a paper. For example, former drug addicts, people engaged in illegal activities, or victims of domestic violence. It could also be because you think that a participant could help you find people more efficiently. For example, if you are looking for expatriates who have been in Prague for at least 25 years. If you find a few, they may have friends that could help with the research. Network sampling also helps to build trust with the researcher. If a recovered drug addict recommends my research to a friend with the same history, it saves time in trying to build up trust for carrying out the research. If John thought the researcher was ok, then so does his buddy.

Stratified sampling

Stratified sampling attempts to make a sample that reflects the sub-groups within a target population. If our school, for example, consists of 50% native speakers of English, 30% non-native speakers of English, and 20% "ESL still getting language support" - and I ran a study in which the entire sample was made up of native speakers, the study would not be representative of our school community. A stratified sample of 40 participants should be made up of 20 native speakers, 12 non-native speakers, and 8 ESL students who are still receiving support. Ideally, the participants are randomly selected from each "stratum." The problem with stratified samples is that one has to be careful about how they label the groups within a community. If we were to divide our school into "North American", "European" and "Other" - would the "other" label fairly represent the non-US/Euro population at our school?

For qualitative research- Representational generalization: Candidates may address the extent to which the sample may be representative of ecstasy users in the population from which the sample was drawn. • Inferential generalization: Generalization in qualitative research is not based on statistical representation as in quantitative research. The study in the stimulus question is based on a rather small sample - that is, the participants are selected based on the salient characteristics of having experience with drug use. If the data is sufficiently rich and perhaps corroborated by other similar studies it may be possible to generalize the findings to similar populations. • Transferability would look at how specific variables that define the study could influence the ability to generalize the findings to other similar populations. • Theoretical generalization: Concepts developed in the study or findings of the study could be used to develop further theory. Discussions of statistical generalization is not relevant to qualitative research and should be allocated marks in the lowest markband.

Participants should not be identified, either by name or by descriptive details that would lead to recognition in their community, potentially leading to negative outcomes.

Justice & equitable treatment

Unjust treatment includes publishing an idea that leads to prejudice against a group or withholding treatment that you believe is beneficial from some participants so that you can use them as controls.

Sound & valid methodology

When research findings are publicized, people are likely to take them as fact and policies may be based on them. It is important that the limitations of the studies are explicitly communicated to all constituents.

Deception

It is important that findings are stated precisely and cautiously to avoid misunderstandings or misinterpretations of the research.  The interpretation of the findings should not be overstated. Researchers should also avoid such unethical practices as p-hacking -  the misuse of data analysis by performing many statistical tests on the data and only reporting those that come back with significant results. 

Any misrepresentations of the findings by policymakers or the media should be directly addressed by the researchers.

Reflexivity

Researchers should be aware of how their findings may be used by others. They should make explicit the assumptions underlying their research so that the public can consider whether they agree with these. They should be aware of their own values and biases and how they may have influenced their findings.

Transparency

Researchers also need to be transparent about the funding of their research. It is important that researchers reflect on why the funders have funded their research and how the funders may use any findings of the research.

1.  A psychologist wants to study the effect of loud noise on a person’s ability to concentrate. In order to do this, he blasts loud music through the participants’ earphones while they are trying to read a text.

The level of the noise could be a concern. The researcher needs to make sure that the noise does not cause any damage. The participants would also have to be informed of the nature of the experiment before they begin. If they find it stressful, they have the right to withdraw.

2. A criminal psychologist wants to carry out an observation of how prostitutes convince someone to have sex. He will do this by a covert observation, where he poses as a person wanting her services.

Deception is a key problem here. He is dishonestly approaching the prostitutes as if he were a customer. In addition, there is the question of researching something that is "illegal." Prostitution is illegal in many countries. So, the identities of the individuals must be protected. Though, this raises the question of knowing that someone is doing criminal activity but not turning them in.

3. A developmental psychologist wants to test the problem-solving abilities of elderly people residing in a retirement home. He goes from room to room in the center and asks them if they would be willing to do a Suduko puzzle.

It appears that consent is being obtained, but the consent is not really "informed." It is also questionable how healthy the individuals are in this study. It would be a better study if the participant worked together with the administration of the home in order to gain participants. However, that consideration is not an ethical consideration. Overall, there is nothing out of the norm here for ethical considerations.

4. In order to determine the role of vasopressin, a hormone, on the sexual behavior of animals, DNA that promotes the production of vasopressin is grafted onto the DNA of animals that lack it.

Animal research in and of itself is not an ethical consideration - it is a sampling consideration. However, in this case, the researchers are looking to permanently alter the physiology of the animals. This could lead to irreparable harm. This is a serious ethical consideration.

5. A researcher wants to carry out a series of interviews with a group of Afghan War veterans to assess the extent to which personal experience in the war has affected their perception of the future.

There is nothing unethical about this study. The researchers would have to gain informed consent, guarantee the confidentiality of the data and allow the participant the right to withdraw their data. However, it would be important that the interviewers are carefully trained so as to not cause any undue stress for the veterans.

6. A psychologist hopes to explore the motivation of students for academic dishonesty – for example, plagiarism or cheating on an exam. In order to do this, he hopes to obtain student records from the head of the school.

This is a rather problematic study. Obtaining student records could be considered unethical. Schools in different countries have different laws about the release of school records. It is probable that consent would have to be obtained from the families involved. If not, then the data should be anonymized so that the identities of the individuals are kept confidential.

7. A psychologist wants to study male sexuality with a sample of university students. He is planning on using a focus group in which they will discuss their earliest sexual thoughts, and whether they have ever had homosexual fantasies. In order to guarantee the accuracy of the data, the focus group will be recorded.

This study is highly problematic. First, informed consent would have to be obtained where the individuals were fully aware of the topic of the focus group. Secondly, there would be the problem of confidentiality. There is no real guarantee that information from the interview won't be shared outside of the room by members of the focus group. This is an ethical concern, but it may also mean that no one wants to share openly in the group. Finally, recording the data is concerning. The data would have to be destroyed in order to make sure that the confidentiality of the individuals is protected.

8. A school psychologist wants to try out a new approach to reading for 5-year-olds. The plan is to have one class have this new special program and the other class would continue with the former program. The goal is to see if the group with the new program develops higher reading skills than those without it.

Studies of this nature are always problematic. First, there is the question of consent. Can the school give consent, or should the parents of the children? Usually, it is the parents of the students that most give consent for such types of experiments. Secondly, if the new program actually helps the students, then the results seem to justify the experiment. However, if they actually lead to a decrease in ability, this could have a long-term and potentially irreversible effect on the children's ability to read.

9. A child psychologist wants to test the effect of television violence on children’s behavior. She will show a group of 6-year-old children a movie in which a murderer hunts down and kills small children. She will then have the children work in pairs on a frustrating puzzle to see if they exhibit any aggression toward each other.

There are several considerations here. First, parental consent would have to be obtained. Secondly, there is the question of undue stress or harm caused by watching the films. The nature of the films may lead to long-term nightmares or social-avoidant behavior. Finally, the goal is to induce aggression. There is a question of whether these behaviors would be reversible.

• Paper 3 is an HL-only paper.

• The paper accounts for 20% of the final grade.

• Students have 1 hour to answer the paper.

• Students answer three short answer questions on approaches to research based on an unseen study. 

A typical Paper 3 may look the following way:

Question 1 will always look the following way:

Identify the method used and outline two characteristics of the method. [3]

Describe the sampling method used in the study. [3]

Suggest an alternative or additional research method giving one reason for your choice. [3]

Question 2 will always ask you about the ethics in context of the study scenario e.g.
 

Describe the ethical considerations in reporting the results and explain ethical considerations that could be taken into account when applying the findings of the study. [6]

Question 3 will always ask about one of three issues: Generalisability, Credibility, or Bias. You may obtain the following questions:

Discuss the possibility of generalising the findings of the study. [9]

Discuss how a researcher could ensure that the results of the study are credible. [9]

Discuss how the researcher in the study could avoid bias. [9]