Research Methods in Psychology – Q&A (PSYC 304)
Unlock all answers in this set
Unlock answersquestion
Discuss the balance between potential benefit of a research project and its potential costs to research participants
answer
If harm is to be done to participants, the results of the study must have a considerable potential to improve society at large. Certainly there are many useful things that could be garnered and justified from participants subjected to moderate pain or discomfort, but many ethical issues become relevant when asking participants to pay higher costs
question
What is informed consent? What is the importance of informed consent in the ethical use of research participants in experimentation?
answer
Informed consent is explaining the purposes and procedure of the experiment to the participant in terms understandable to them, and confirming that they understand and are willing to participate either in writing or verbally (best if recorded).
question
What is the researcher's responsibility with respect to the research participant's right to privacy?
answer
Researchers must code sensitive data, respect people's right not to reveal certain pieces of information about themselves, and keep in mind that there is much potential value in information that people may be reluctant to reveal.
question
What is deception? Why is deception necessary in some research? Under what circumstances is deception acceptable in research?
answer
Deception involves lying or misleading participants regarding the actual purpose or hypothesis of the experiment. Often it is necessary because knowing the true nature of the research would change its outcome. It is acceptable only if the prospective scientific, educational, or applied value of the study overrides the possible psychological damage of the deception and effective non-deceptive alternative procedures are not feasible.
question
What is debriefing in research? What functions are served by debriefing?
answer
Debriefing is the informing of participants of the true nature of an experiment that has involved deception. This reduces stress and damage that may have been involved in deception and ensures that participants experience as much personal and educational value as possible from the experiment.
question
Evaluate Milgram's (1963) obedience research in relation to the ethical principles discussed in this unit.
answer
Milgram's deception caused the participants to feel grief, horror, immense stress, and even led one participant to have a seizure. Debriefing led to participants likely feeling worse about themselves, though relieved that the 'subject' was only an actor and was still alive. However the immense importance of this study could be argued to be worth the psychological trauma the participants endured.
question
Briefly compare the codes of ethics of the Canadian Psychological Association and the American Psychological Association.
answer
...
question
What are the views regarding the role of participants in psychological research?
answer
Some encourage a strict divide between researcher and participant or subject, whereas some view participants as collaborators and part owners of the data which they have provided.
question
How are animals protected against inhumane treatment as research subjects in accordance with ethical guidelines for psychologists?
answer
In Canada, the Canadian Council on Animal Care (1993) has provided ethical guidelines for the treatment of research animals. These guidelines are to be followed by researchers affiliated with Athabasca University, as stated in the University's Animal Care Policy (2004). Standards are set for laboratory animal facilities, laboratory animal care, experimental animal surgical procedure, the use of anesthesia, and euthanasia. Researchers using animals are supervised and trained, smaller numbers of animals are used if possible, and all work is done in consultation with a veterinarian to reduce pain and suffering.
question
How might these ethical guidelines affect the validity of research?
answer
Smaller numbers may introduce statistical issues, certain types of important research may not be able to be done, and warm feelings between researchers and their animal participants could bias results.
question
What distinguishes non-experimental from experimental research
answer
The distinction between experimental research and nonexperimental research is based on the degree of control that the researcher has over the subjects and conditions of the research - manipulation and assignment versus observation. Non-experimental research's major purposes are to determine correlations, seek qualitative data regarding how people articulate their opinions or subjective experiences, gather support for a relationship or law predicted by experimental research, and gather preliminary information that can be used to develop hypotheses that can be tested in experimental research. In order to reach these ends, observational researchers observing ongoing behaviors, archival researchers examine existing records, case studies are created and analyzed, and surveys inquiring about the relevant variables are conducted.
question
Why is non-experimental research often called correlational research
answer
Non-experimental research is often called correlational research because, as far as causation goes, can only determine correlation, not causation, as would be more likely in experimental research. The goals of correlational research are to describe and to predict. In correlational research, the researcher can determine a relationship between two variables, but cannot determine which variable is the cause and which variable is the effect. Often, two variables are correlated to each other because both are correlated to a third, common variable. In such cases, the correlation between these two variables is called a spurious relationship. In general, correlation does not imply causation. However, correlations between variables can be good indicators of potential causes of behaviour.
question
Briefly describe several common varieties of non-experimental research
answer
Non-experimental research can be conducted in several common ways: observation, archival research, case studies, and surveys. In observation, the researcher simply observes an ongoing behaviour of interest (e.g., pedestrians crossing the street). In archival research, the researcher examines existing records kept by individuals or institutions (e.g., student registration records kept by a university). In case studies, the researcher takes advantage of a relatively rare opportunity to examine an individual, an institution, or a real-world event intensively (e.g., the victims' responses to a natural disaster). In surveys, the researcher asks people questions, often in the form of a questionnaire (e.g., Canadians' attitudes toward same-sex marriage).
question
How is naturalistic observation different from the casual observation that most people do
answer
Naturalistic observation involves observing behaviours in a natural setting without any attempt by the observer to intervene or to influence the observation process. A natural setting is where behaviour ordinarily occurs; for example, a playground, a shopping mall, or a bus terminal. As the textbook points out, it is for this reason that naturalistic observation is often called unobtrusive research. Naturalistic observation is also called non-reactive research because people are unaware that they are being observed and, as a result, they do not react to the presence of the observer. Most people would not be as precise, thorough or empirical in their observations of others.
question
Describe briefly three common ways of sampling behaviours in observational research
answer
To describe behaviours accurately, a researcher must observe a sample of behaviour that is representative of the larger population of all possible behaviours. Researchers use three procedures to sample behaviours: time sampling, event sampling, and situation sampling. Time sampling refers to selecting various time intervals, either systematically (e.g., every 15 minutes) or randomly, for their observations. Often, researchers will schedule observation intervals systematically, but will observe at random times within an interval (e.g., at a random time within every one-hour interval). Event sampling refers to recording each event that meets a predetermined definition (e.g., fighting), which tends to happen infrequently. Situation sampling refers to observing behaviour in as many different locations (e.g., in school playground and at home) and under as many different circumstances and conditions as possible.
question
When is physical trace research useful? Identify and describe the two common forms of physical traces, and give an example of each.
answer
Apart from observing behaviours as they are occurring, researchers can also observe physical traces, which are physical evidence of past behaviour. Physical traces are of two forms: use traces and products (Shaughnessy et al., 2009). Use traces are the physical evidence that results from use (wear and tear); for example, the bare spots on a lawn indicating a short-cut from one building to another. Products are the physical evidence of what is left behind from past behaviour, for example, empty beer bottles left after a party. As an unobtrusive measure of behaviour, physical traces have the advantage of being a non-reactive measure. Physical traces are often collected indirectly, after the people whose behaviours are being measured have left the scene, which makes the measurement of physical traces non-reactive. In other words, people's behaviours are not influenced by the act of measurement. Researchers can use physical traces in creative ways to test hypotheses about behaviours and preferences. These measures can be used to assess the validity of certain kinds of verbal reports. Consider, for example, the posters of pop music stars on the walls of a teenager's room and the teenager's expressed liking for his or her favourite stars.
question
Propose a hypothesis that may be tested using physical traces
answer
- What types of alcohol do people consume at a frat party? - Approximately how many cigarettes are smoked outside of a large apartment building?
question
What are the advantages and disadvantages of using physical traces as measures to test hypotheses about people's past behaviour
answer
Other factors can influence the deposit or erosion of physical traces, so it is important that physical trace measures not be used alone. The researcher must consider alternative hypotheses or explanations, and then collect data from other sources to refute these alternative hypotheses. The best approach is to use multiple methods to collect data to test the same hypothesis. If different methods lead to the same conclusion, then the researcher has increased confidence in the hypothesis One can't be entirely sure that the traces are left by the target population. They are useful in that research is non-reactive and objectively verifiable
question
Under what conditions is participant-observer research useful
answer
participant observation involves an observer taking part in a group's activities, and observing and recording the behaviours of people in the group under study. The researcher is often interested in understanding the behaviours or activities of the group from an insider's perspective. This method is used when a researcher wants to study a small, often isolated group in which little is known about the group, or when the group's activities are not generally accessible to the public. For example, a researcher interested in the behaviours of gang members or the group dynamics of a gang would likely use participant observation.
question
Describe the ethical problems that participant-observer research poses
answer
Lack of consent- Invasion of privacy- Sometimes researchers use deception- Information released about the group could be damaging to it
question
What are the advantages and disadvantages in conducting archival research
answer
The most obvious advantage is that it saves time in collecting data because the data have already been collected. Another advantage is that as the data are often a matter of public record (kept by government agencies or the media), reducing the researcher's ethical concerns (Shaughnessy et al., 2009).Disadvantages of using archival data in research are of five main types. First, inherent biases may be present in archival data because the data are collected by others for a particular purpose, which means they are not necessarily designed to address the researcher's specific question. Second, only some records are produced and kept (called selective deposits) and some records are missing and incomplete (called selective survival). Third, the quality of data may not be as good as a researcher anticipates, in that there may be errors in record keeping. Fourth, changes may have occurred in the way records are kept or in the definition of categories (e.g., types of crime). Finally, reactivity may have been a problem when the archival record was produced (Shaughnessy et al., 2009). When conducting any research, a researcher needs to identify alternative explanations for the research results and gather additional evidence from independent sources to refute any alternative hypotheses.
question
Outline the steps in conducting content analysis. Illustrate the steps, using your own example
answer
Archival data in the form of textual or graphic materials can be analysed for their content (e.g., newspaper articles or television commercials). Shaughnessy et al. (2009) outlines three basic steps in carrying out a content analysis. The first step is to identify a relevant archival source. The second step is to select a representative sample from the source. The last step is to code; that is, to define relevant descriptive categories and units of measure.One approach is to code the manifest content; that is, counting the frequency of some objective measure. An example of this approach would be to count the number of column inches devoted to a particular topic in a newspaper. Another approach would be to code the latent content; that is, recording the themes as interpreted by the researcher. It is useful to compare the results of both the manifest and the latent content analysis. The researcher should use at least two coders to establish the inter-rater reliability of the codes. To eliminate coder biases, coders should be blind to important aspects of the study, such as the main hypothesesI want to study the level of interest newspapers have in covering protest groups. First I would go to a library and obtain all major national and city newspapers. From this group, I would randomly sample as many editions as were feasible given the amount of time and staff available. I could measure the space given or words typed, and what page the story was on. A minimum of 2 coders should be used to ensure accurate data.
question
What problems are associated with analysing archival data
answer
Incomplete, inconsistent, or biased sources of information. Experimenter bias, etc.
question
Outline the characteristics of case studies. a case study as an intensive description and analysis of an instance, or case, of a phenomenon, such as an individual, an institution, a treatment, or an event.
answer
Researchers take advantage of the occurrence of a real-life event to examine a contemporary phenomenon using data from multiple sources. Case studies have two advantages. First, they can provide insight into possible causes of people's behaviour that can be tested later with more controlled research approaches. Second, case studies allow for the study of rare events. Case studies also have disadvantages. First, researchers can rarely infer any cause-effect relationships from the results because of the presence of extraneous variables. Second, subjective observer biases are likely to occur. Third, it is difficult to generalize the results to other cases
question
What is the difference between research method and research procedure
answer
Method is a broader term that encompasses all aspects of the study, including the logic of the design and the steps for carrying it out. Procedure refers only to the latter - what the researcher does in translating the design into action.
question
Why is it necessary to conduct a pilot study
answer
Once a protocol has been developed, a pilot study will be necessary to find the bugs in the procedure. A protocol is a list of all the steps that a subject goes through in a study Most studies done without a pilot have disappointing results, and researchers with reputations for excellent design are often those who do extensive pilot research.
question
For what purposes are surveys used & What are the characteristics of surveys
answer
Surveys are widely used to describe people's thoughts, feelings, opinions, attitudes, and preferences. They are also used to predict people's behaviours through the examination of relationships between variables. Shaughnessy, Zechmeister, and Zechmeister (2009) point out that surveys have several characteristics. First, surveys generally involve sampling. Second, surveys involve the self-reporting of behaviours, feelings, thoughts, and preferences, either in oral or written form. Third, all respondents answer the same set of predetermined questions to allow for comparison of responses.
question
Distinguish among population, sampling frame, sample, and elements. Give an example to illustrate these distinctions
answer
Sampling involves selecting a sample (i.e., a subset of individuals) from the population (i.e., all individuals of interest to the researcher). Before a researcher selects a sample, he or she must develop a sampling frame; that is, a specific list of the individuals in the population. Each individual who falls within the sampling frame is called an element
question
What is a representative sample & Why is it important to select a representative sample
answer
The extent to which the results of a sample are generalizable to the population depends on the degree to which the sample is representative of the population. A sample is representative of the population when the sample has the same distribution of characteristics as the population. When some segments of the population are over-represented in the sample, selection bias results.
question
What are the relative advantages and disadvantages of mail surveys, personal interviews, telephone surveys, and Internet surveys
answer
Mail • low cost • no interviewer bias• best for personal, sensitive topics (where anonymity is preserved) • questions must be self-explanatory• little control over order of questions answered • response bias (e.g., low response or return rate) Personal Interview • interviewer can motivate interviewees, clarify questions, and monitor order of questions answered• higher response or return rate than in mail surveys• possible to use visual aid • interviewer bias• much more expensive than mail or telephone surveys• difficulty reaching interviewees• potential interviewer effect (respondents say what they think the interviewer wants to hear)• supervision of interviewers more difficult Telephone • easier access to respondents• quick completion• tighter supervision of interviewers• low cost • possibility of using computer-assisted interviews • possible selection bias• only suitable for short, straightforward surveys• possible interviewer bias• telemarketing and donation solicitation, and option to screen calls and to use voice mail have reduced accessibility• impossible to use visual aids• less anonymous than mail or web-based surveys Internet • efficient: low cost and speedy• potential to reach a large number of respondents of diverse backgrounds and from difficult-to-access populations• can control sequence and branching of questions• can check for invalid responses easily and consistently • response bias, just as in other methods• selection bias: e.g., accessibility to the Internet, website for special-interest groups• difficulty in controlling how the survey is completed, e.g., alone or in groups, clarity of instructions • difficulty in checking variables about respondent (e.g., gender, age) that are on interest to the researcher
question
What are the relative advantages and disadvantages of using open-ended and closed-ended questions in a survey
answer
Open-ended Advantages • More complete answers that may provide reasons for certain attitudes or behaviours• More likely to discover something unexpected by the researcher Disadvantages • More difficult to record and score responses• Requires more effort from respondents• Difficult for less-articulate respondents to answer • Closed-ended • Easier to code and analyze • Requires less effort from respondents; no need to articulate answers Disadvantages: Reduces expressiveness and spontaneity• Respondents may have to choose a less-than-preferred response because no alternatives really capture the respondent's view• More difficult to discover errors in coding or misinterpretation of questions
question
State the criteria for a useful questionnaire item
answer
Each item should address a single question in a clear, unambiguous manner; avoid biasing language; write closed-ended questions that are mutually exclusive and exhaustive so that alternatives are clear; word questions so that each alternative appears equally socially desireable; try not to format the questionnaire with too many binary closed-ended questions as they encourage acquiescence; determine type of question (true/false, multiple-choice, ratings)
question
Describe how the ordering of questions in a survey may influence the quality of data
answer
For self-administered questionnaires, such as a mail survey, it is advisable to ask first those questions that are of greatest interest to the respondents so as to capture their attention, and to leave the questions about demographic data at the end. In personal or telephone interviews, it is important for the interviewer to establish rapport with the respondent. Such interviews often begin with questions that are easy to answer, such as demographic questions (Shaughnessy et al., 2009).The order in which particular questions are asked can influence the answers given by respondents. The generally accepted method is to begin with the most general question and then move on to more specific questions within a general topic. Often, branching items are used to direct respondents to skip inappropriate questions. The use of branching items reduces the demand on the respondent's time.
question
Under what circumstances will responses be biased
answer
Results will be biased when certain wordings are used (i.e., making one position look more socially desirable than the other), or when the general context of the survey is different. Certain sequences of items will encourage different results. Limited sampling will lead to biased results
question
What is acquiescence& How can researchers identify acquiescence
answer
acquiescence: the tendency to agree with a statement on a questionnaire, regardless of its content. A researcher can ask the opposite question and see if that response is affirmative as well.
question
What is the response rate or return rate of a survey & What can researchers do to increase the response rate of a survey, particularly a mail survey
answer
A response or return rate is the percentage of individuals in the sample who return the completed survey. Drop off or mail administered surveys generally have very low response rates, often less than 50%. Keep it simple, short and interesting. Putting relevant questions first and demographic information last will make it more appealing. Giving a prize or money incentive for a response would likely increase return rate.
question
Discuss two limitations of survey research
answer
Social desireability bias and low return rates.
question
Briefly describe the ethical issues in conducting survey research. What steps should researchers take to ensure that ethical principles are followed
answer
Two important and related ethical issues are particularly relevant in survey research: anonymity and confidentiality. Many surveys are completed anonymously; that is, without any identification of the respondent. A researcher who needs to track whether a person has responded, or who needs to re-survey the same person over a period of time, can use some means of identification (e.g., a number or a code). In such cases, the researcher must destroy any information that could associate a respondent with participation in the study or with the respondent's data.When participation in a survey is not anonymous, the researcher must assure confidentiality of respondents' participation in the survey. The data must be accessible only to those in the research team. In addition, the researcher should report the survey results only in aggregate, summarized form so that a particular respondent's participation or responses cannot be identified. Assurance of confidentiality is particularly important when the survey relates to sensitive topics, such as sexual activities and substance abuse.Informed consent is another important ethical guideline to be followed. A researcher must let respondents know that their participation is voluntary, that they can choose not to answer any of the questions in the survey, and that they can terminate their participation at any time, without any penalty.
question
Name several prominant psychologists in history who used single subjects in psychological research. Briefly describe their areas of research
answer
Gustav Fechner, who some historians say is the founder of psychophysics, invented the basic psychophysical methods that are still used to measure sensory thresholds and discovered principles of psychophysics still taken seriously today. Inspired by Fechner's work, Hermann Ebbinghaus used similar methods to study memory, and I. P. Pavlov in conditioning of individual dogs
question
Compare the assumptions between the single-subject approach and the individual-differences, group-research approach to psychological research. What are the advantages of the single-subject approach
answer
Typically, the single-subject design focuses on an examination of behaviour change in one person. One obvious advantage is that this allows for an intensive study of the individual client and experimental evaluation of treatment for that particular client. A second advantage is that it requires only a small number of subjects. Where experimental subjects are difficult to find (e.g., patients with damage in a particular part of the brain), this may be the only feasible design. The third advantage is that withholding a potentially beneficial treatment from some patients is not required in order to provide a control group. An experiment that implies large groups of participants will be likely to discover that an independent variable has an effect even if the effect is a minor one. Thus the discovery might have no practical useavoiding the ethical issue that some participants will be placed in a control group and will not receive treatment or receive inferior or delayed treatment. Also, if a researcher cannot locate enough participants, a single subject approach will have to suffice. Flexibility in design: an experimental group of subjects must be designed so that all subjects received the same experience can be compared. During an experiment on behavior modification, and experimenter may discover that a subject does not respond to reinforcer that has worked on previous subjects. In a single subject design, the experiment can be modified on the spot by switching reinforcers or altering the instructions.
question
What are the disadvantages of the single-subject approach
answer
1st, some effects are small relative to the amount of variability in the situation. It may be impossible to control the other sources of variability sufficiently to observe the experimental effect in one subject. This is why modern statistical methods were developed in the 1st place. 2nd, some experimental effects are by definition between subjects effects. It is impossible to have a participant who simultaneously receives 2 opposite sets of instructions in a social psychology experiment, or who is taught the same material by 2 different methods.
question
What are the common single-subject research designs
answer
Some common single subject designs include:comparison or AB designs: when one simply measures the baseline behavior (A) and then introduces the treatment (B) withdrawal of treatment or ABA design: research design that includes a baseline period, and treatment period, and subsequent withdrawal of treatment. The argument that the treatment is because of the changes considerably strengthened if the treatment is withdrawn after a period of time research design that includes a baseline. And the behavior shows the returns were the baseline. Repeating treatments (ABAB designs): an ABA design with treatment repeated after the withdrawal phase. Repeated presentation and withdrawal of a variable can produce strong evidence for the validity of the independent variables effect
question
What is the ABA design & What two principal problems are associated with this design
answer
The most basic single-subject design is the ABA design, also called the reversal design. First, the baseline (A) level of a behaviour is established (i.e., the subject's level of performance prior to introducing an experimental condition). Next, the change (B) is introduced and maintained while performance is measured. The final phase involves returning to the A condition (the baseline phase). Such a design has substantial internal validity. First, each participant uses his or her own control as the baseline data serve as a reference level against which to compare the data collected during the experimental phase. Second, the requirement that the behaviour be reversible following the withdrawal of the experimental condition rules out the possibility that changes from the baseline level are caused by extraneous variables. In practice, increasing or decreasing baseline trends, or excessive variability in behavioral measures can make it difficult to evaluate the effects of an experimental condition. If the results of a single-subject experiment can be replicated with additional subjects, both the internal validity and external validity of the study would be increased
question
When is the ABAB design superior to the ABA design
answer
This design is superior when one wishes to continue the beneficial treatment (B).
question
What is a multiple-baseline design & Under what conditions is this design useful
answer
Multiple baseline design is a research design that introduces experimental manipulation at different times for different behaviors to see if behavior change coincides with manipulation. For example, a researcher trying to determine if rewarding a child for doing certain personal tasks is effective could begin by rewarding toothbrushing, face washing, handwashing, and hair combing all at the same time, but it is possible that the experimenter, the attention received, or a spontaneous decision could be responsible for the change. However the researcher could increase internal validity by rewarding only toothbrushing the first week, facewashing the second, and so forth to determine whether or not the reward is what is indeed responsible for the behavior change.
question
What is a changing-criterion design & Under what conditions is this design useful
answer
Research design that introduces successively more stringent criteria for reinforcement to see if behavior change coincides with the changing criteria. Like the multiple baselines design, a changing criterion design is useful when the behavior change irreversible or when a return to the initial baseline is not desirable.
question
Describe briefly two areas of psychology in which the single-subject approach is commonly used
answer
Clinical applications of operant conditioning principles and psychophysical experiments still use the single subject research design. Psychophysical examples include visual and audio thresholds.
question
Define true experiment
answer
In a true experiment, the experimenter has complete control over the experiment: independent variables are manipulated by the experimenter, research subjects are randomly assigned to different experimental groups and the control group. Extraneous variables are controlled so that any observed effects can be attributed to the independent variable.
question
Define and distinguish the terms factor, level, and condition when referring to independent variables
answer
A factor is an independent variable in an experiment, i.e., new medication. A level in an experiment refers to a particular value of an independent variable, i.e., taking 3 pills/day. A condition is a group or treatment in an experiment, i.e., 4 pill/day experimental condition.
question
Describe the two basic elements of good experimental design
answer
A control group or condition and random allocation of subjects to groups will provide control over so many different threats to validity that they are vital for a valid experiment.
question
What are within-subjects designs & What is the basic strategy for achieving control
answer
In a within-subjects design, each subject acts as their own control group. Ordinarily , researchers will avoid within-subjects designs if they believe that order and sequence effects will be substantial. The basic method for controlling for order and sequence effects is called counterbalancing. This is done by arranging that subjects experience the various conditions in different orders. Each condition should follow every other condition equally as often.
question
What are order effects and sequence effects & How are they different
answer
Order effects are changes in the subject's performance resulting from the position in which the condition appears in an experiment. Sequence effects are changes in the subject's performance resulting from interactions among the conditions themselves.
question
How do you control for order and sequence effects within subjects
answer
Randomization can be used when each condition is given several times 2 each subject or when a sufficient number of subjects will be tested such that one particular sequence is unlikely to have much influence on the outcome. Block randomization is the control procedure in which the order of conditions is randomized but with each condition being presented once before any condition is repeated. Reverse counterbalancing is a method of controlling which conditions are presented in order the 1st time and then in reverse order.
question
When is reverse counterbalancing ineffective in controlling for order effect
answer
If there is a practice effect then counterbalancing will not be effective in eliminating the order effect. One way of improving such an experiment is to provide enough practice beforehand that the practice effect is eliminated.
question
How do you control for order and sequence effects within groups
answer
If presenting each condition enough times to randomize the order is not possible, or if counterbalancing within subjects is not appropriate, you must leave order and sequence confounded with conditions within subjects. Then you must control for ordering sequence across subjects, essentially within the group.
question
What is complete counterbalancing, What does it control for, What is its greatest disadvantage
answer
Complete counterbalancing within a group occurs when each condition occurs an equal number of times in each rank order position, and follows every other condition an equal number of times. Thus you control for ordering sequence within a group of subjects, even though every subject experiences a biased sequence. The disadvantage of this method of counterbalancing is that as the number of conditions increases the number of orders required increases geometrically - meaning 6 orders of 3 conditions, 24 orders of 4 conditions, and 120 orders of 5 conditions, meaning you need at least as many subjects.
question
What is the Latin Square technique and When is this design preferred over complete counterbalancing
answer
You can control for the order in which each condition occurs with your subjects with many fewer subjects than required by complete counterbalancing if you give up the requirement that each condition follow every other condition in equal number of times. You would be controlling for order, but not for sequence of conditions. This type of incomplete counterbalancing is called the Latin square technique, after the ancient puzzle of finding ways to arrange a number of letters in the matrix such that each letter occurs only once in each row and column. If the letters represent conditions, the columns represent order, and the rows represent subjects, you're controlling for order effects with the Latin square counterbalancing technique. The advantage of the land where technique over complete counterbalancing is that it permits greater flexibility in choosing the number of subjects to be tested; instead of needing 24 or 48 subjects, you can use only 4 or 8.
question
What are between-subject designs and Explain the basic strategy for achieving control
answer
A between subjects design is one in which each subject is exposed to one level of the independent variable only. Groups are compared to other groups with varying levels of treatment. The basic strategy for achieving control is the inclusion of a control group or groups that are not exposed to the independent variable and/or varying levels of the independent variables
question
When are between-subjects designs preferred over within-subjects designs
answer
Sometimes carryover effects, or the need for naivety will make a between subjects design necessary.
question
Describe three designs to avoid in research, and give reasons for avoiding them
answer
One group posttest only design: research design that measures the behavior of a single group of subjects after they are given a treatment. This is invalid because there is no baseline measurement. Also there are no different levels of treatment, so there is no evidence that the treatment itself is what exactly is causing a change. This is basically what happens and people recommend products based on their own experience. Post test only design with non-equivalent control groups: a non-equivalent control group is one in which the subjects are not randomly selected from the same population the experimental group. One group pretest posttest design: research design that measures the behavior of a single group of subjects both before and after treatment. Other potential causes of change are not controlled for.
question
What are factorial designs
answer
A factorial design is one in which two or more independent variables (or factors) are manipulated simultaneously to determine their separate and combined effects on the dependent variable. For example, one may be interested in testing the effect of amount of sleep and noise level on typing performance. If the first factor has two levels (e.g., four hours versus eight hours of sleep) and the second factor has three levels (e.g., high, medium, or low noise levels), then there are six possible combinations of the two factors. Research participants randomly assigned to a treatment condition (or cell) would receive the designated combination of the two factors (e.g., four hours of sleep and high noise level).
question
What are the advantages of using factorial designs
answer
factorial designs produce greater precision than do single-factor designs. By examining the interaction effects of more than one factor on the dependent variable, the experimenter can appreciate that a behaviour has many causes that interact in a complex way. Another advantage over single-factor designs is that a potentially confounding variable can be built into the design by including that confounding variable as a factor in a factorial design.
question
How many cells (combinations) are there in a 2 x 3 x 4 factorial design
answer
24
question
What are main effects
answer
A main effect in a factorial experiment is the effect of one independent variable, averaged over all levels of another independent variable.
question
What are interaction effects and How can one tell graphically if there is an interaction between two variables
answer
When the effect of one independent variable depends on the level of another independent variable. The dependent variable is plotted as a function of an independent variable, with the other independent variable as a parameter. If the graphical representation of a factorial experiment shows curves that are not parallel, there is an interaction between the variables. If they are parallel, however, there is no interaction. This is true no matter how complicated the curve may be, however small differences may be the result of chance, so statistical evaluation may be necessary.
question
What are the common types of interactions in psychological research
answer
An antagonistic interaction: interaction in which the two independent variables tend to reverse each other's effects. Synergistic interaction: Interaction in which the two independent variables reinforce each other's effects.Ceiling-effect interaction: Interaction in which one variable has a smaller effect when paired with higher levels of a second variable.
question
What is the difference among within-subjects, between-subjects, and mixed factorial designs
answer
The within-subjects factorial design requires the fewest subjects to achieve a particular degree of power, the mixed design the next fewest, and between-subjects design the most. Within subjects is good if there are no or negligible carryover effects.
question
How are quasi experiments different from true experiments
answer
a true experiment must generally satisfy three conditions: (1) the groups assigned to different treatment conditions must be initially equivalent, which is usually accomplished by random assignment of subjects; (2) the independent variable must be a manipulated variable; and (3) the existence of a control condition against which the behaviour of experimental subjects (as in between-subject designs) or of the same subjects in the treatment condition (as in within-subject designs) is compared.Where one or more of these conditions are not satisfied but the purpose of the research is to imply a causal connection between the independent variable (i.e., a selected subject variable or a natural treatment) and a subsequent behaviour, we have a quasi experiment. Thus, a quasi experiment is one that resembles an experiment but lacks at least one of its defining characteristics. In true experiments we manipulate variables, whereas in quasi experiments we observe categories of subjects.'
question
Under what circumstances can quasi experiments be a useful tool
answer
One good reason for conducting quasi experiments is that it may be the only possible way to study the impact of a natural event, for example, how residents cope after a flood. For ethical and practical reasons, the researcher cannot manipulate the occurrence of such an event and cannot randomly assign residents to flood versus no flood groups. In cases where the impact of a treatment program or intervention is introduced, the researcher is seldom able to assign clients randomly to treatment versus control groups. The research data collected in quasi experiments can serve as a rich source of testable hypotheses to be tested in a carefully controlled laboratory experiment. Conversely, the external validity of laboratory findings can sometimes be assessed using the quasi-experimental approach.
question
What is the major weakness of quasi experiments
answer
The lack of control, either in applying the independent variable or randomly assigning subjects to control & experimental groups
question
What is a non-equivalent control group design with pretest and posttest? Under what circumstances are the results of quasi experiments that use this design interpretable?
answer
A non-equivalent control group is a group of subjets that is not randomly selected from the same population as the experimental group. any behavioral change from pretest to posttest of an experimental group is compared with that of a non-equivalent control group. Two result patterns of this design are generally interpretable: (1) if the behaviour of both groups are the same on the pretest and the experimental group's behaviour has changed on the posttest while the control group's has not, or (2) if there is a crossover between the experimental group and the control group on the pretest and on the posttest (i.e., the experimental group has a lower measure than the control group on the pretest but a higher measure on the posttest).
question
What is an interrupted time-series design & Under what circumstances is this design useful
answer
The interrupted time-series design comprises three phases: (1) a series of pretest measurements over time, (2) the occurrence of the significant event (treatment or intervention), and (3) a series of posttest measurements. Many different patterns of results can emerge from interrupted time-series studies. In general, any change in either the slope or the interval at the time of treatment or intervention would suggest the intervention to be the cause of behaviour change. Problems with data interpretation, however, arise frequently when: (1) the treatment or intervention is gradual rather than abrupt, (2) there is a delay impact of the intervention, (3) the time series is short, and (4) there are incomplete or inaccurate archival data (Cook & Campbell, 1979).
question
What different strategies are used in non-equivalent control group designs and interrupted time-series designs to control for confounding variables
answer
Ruling out alternative hypotheses, i.e., comparing similar groups outside of the 'study' groups. Taking multiple baseline readings or having access to baseline material from before the independent variable was introduced to rule out maturation, practice, etc.
question
What are repeated-treatment designs? How do these designs control for confounding variables?
answer
Repeated-treatment designs attempt to improve the validity of the experiment by presenting the treatment more than once. The subject's response is measured before and after the introduction of a treatment, then the treatment is withdrawn and the whole process is begun again. The treatment must be one that can be withdrawn without causing complications in the analysis of data. Also, whatever change is found between pretest 1 and posttest 1 should be in the same direction as that between pretest 2 and posttest 2.
question
Which quasi-experimental designs can be used with a single subject (an individual, a company, or some other units)
answer
Repeated treatment & interrupted time-series design can be used with single subjects
question
What are the common quasi-experimental designs used in developmental psychology, What are their relative advantages and disadvantages
answer
Cross-sectional studies test different age groups at the same time. The disadvantage is that people at different ages were all born at different times, so age is confounded with date of birth. A cohort is a group that has some characteristic in common, and so people with this characteristic are treated as a group. To reduce cohort effects, a longitudinal study, which follows one or several cohorts separately, could be used. However, longitudinal studies take years to complete as the cohort ages (practical problem). The main theoretical problem with a longitudinal study Is that it confounds age with time of testing. Secular trends, or changes taking place in the general population over time, can cause confounds, i.e., technology could be changing, causing differences in performance not related to birth year. A cross-sequential design tests individuals from two or more cohorts at twto or more different times ( a blending of cross-sectional and longitudinal) in order to separate developmental, cohort, and secular effects.
question
What is program evaluation & How can quasi-experimental designs be used in evaluations of social programmes
answer
Program evaluation is a set of techniques for determining the effectiveness of a social service program. Evaluators would compare the program with other programs (standard treatments) or outcomes between those who are and are not participating in a program. Program evaluation is of two major types: summative and formative. The goal of summative evaluation is to assess whether a given program is functioning adequately. Its outcome is used to determine whether a program should be cut back, discontinued, maintained, or expanded with additional funding. The objective of formative evaluation is to recommend changes that will improve program functioning. Potential political implications of the research outcome and the impact on a program's stakeholders may cause resistance to evaluation, especially summative evaluation, to be frequent and fierce. One way to minimize resistance is for the evaluator to establish a rapport with the stakeholders early on in the process. When conducting program evaluations, researchers are well-advised to be cognizant of the politics involved every step of the way. Quasi experimental designs are often used in program evaluations, because clients of the program are more likely to be assigned, rather than to be randomly allocated, to groups for comparison. So, clients might be matched with sometone who has note been served by the program on some key variables, and then the clients and non-clients could be compared to see how well the program is working by using some variant of a non-equivalent control group design.
question
In what ways is scientific communication carried out
answer
Conventions and conferences are often the first place where research is reported. At these professional meetings, scientists meet, present, and discuss recent and ongoing research, challenge one another's ideas, and develop networks of researchers working on similar or related projects. Oral presentations and poster presentations are popular forms of communication in these meetings. Informal discussions over the telephone, by fax, electronic mail, postings on electronic networks, online discussion forums and blogs, and online journals are other means of communication among researchers. In addition, visits to other laboratories or facilities occur fairly frequently.The formal channel of communication among researchers is through publication in professional and academic journals.
question
What are the essential features of a good scientific report? (Refers to writing style)
answer
Clarity: say exactly what you mean as directly as possible Brevity: Make every word, phrase, and sentence contribute to the paper. Do not waste the reader's time Felicity: Pleasantness of style. Make the writing lively and graceful.
question
Describe the parts of a scientific paper and the functions of each part.
answer
A typical research article or report in psychology consists of several parts: title, author(s), abstract, introduction, method, results, discussion, and references. The title should indicate the contents of an article. The abstract is a brief summary (maximum of 120 words) of the article's key points. The introduction specifies the problem to be studied and its importance. It also states any hypotheses to be tested and describes the rationale behind any predictions. The method section describes what was done in sufficient detail for other researchers to replicate the study. This section usually includes descriptions of the subjects or research participants, the apparatus or material used, and the data collection procedure. The results section describes the research findings, and is often supported by statistical analyses results, tables, and graphs. Finally, the discussion section usually includes a brief summary of the main results, their relationship to the hypotheses tested, the researcher's conclusions about the results, and a discussion of any weaknesses in the data that warrant qualifications of the conclusions. When suggestions are made for future research, they are usually included in this section. Full titles of reference articles are found at the end of the article.
question
What is plagiarism
answer
Plagiarism is when one presents another's work as one's own.
question
Briefly discuss the human biases to which science is subject
answer
Cultural and societal values often indirectly influence choices of research questions and interpretation of information. The influence of politics on science cannot be overlooked either as funding for scientific research is provided by government, and, increasingly since in the 1990s, by large businesses and industries, whose political forces may well determine whether a particular problem is researched. Research that is considered to be relevant is often pursued in preference to basic research, where the public sees no immediate application. Once a research project is completed, researchers may be pressured to distort the findings to meet with the approval of particular interest groups or stakeholders.
question
Discuss the limitations of science in achieving knowledge
answer
Limited by existing methodology and knowledge from which to base itself on. Science is limited by the state of our technology, and since it is based on empirical observations, it cannot answer all questions. Some hypotheses are untestable because objective data are not available and never will be available. Answers to these questions must be found outside of science, using other ways of knowing. Some hypotheses are potentially testable; although present technology does not permit objective data to be collected, future technology might (Rosenberg & Daly, 1993). For example, the availability of functional magnetic resonance imaging (fMRI) has given researchers in recent years a new tool to examine brain activities while research participants are engaging in specific tasks or behaviours.
question
What are the responsibilities of a scientist
answer
Every scientist has the responsibility to educate the public about his or her work, benefit society by his or her work, and pursue unpopular problems and propose unpopular ideas for the sake of developing new knowledge. And, the very existence of science depends on each scientist's honesty and integrity in reporting his or her work.
question
What is the difference between empirical and non-empirical methods of knowing about behaviour?
answer
Empirical methods are based off of observation and experience, whereas non-empirical methods are based off of non-experiential methods, such as logical deduction or information disseminated by others.
question
When is logic not a good way of knowing the truth?
answer
If it bases it's propositions on a falsehood its statement may be logically valid but untrue.
question
Describe two basic limitations of common sense as a way of knowing.
answer
First, standards of common sense differ from time to time and from place to place according to the attitudes and experiences of the culture (poor external validity). Second, the only criterion common sense recognizes for judging the truth is pragmatism. If a practice works, no systematic attempt is made to see why. Thus, common sense cannot predict new knowledge.
question
How do common sense and science differ?
answer
Science aims to predict new knowledge, can sometimes be counterintuitive, and its postulates should hold true despite time and culture (depending on its type of knowledge, i.e., anthropological knowledge will become outdated).
question
What does \"science is objective\" mean?
answer
This means that science is based on objective, publicly verifiable observations.
question
What are the characteristics of science? Explain.
answer
Observations are empirical and objective in that they are directly observable by another person with normal sensory capabilities. As such, these observations are verifiable and public. Any idea is subject to scrutiny by other scientists and is tentative. Science is self-corrective in that any findings that are not replicated will be discarded and new ideas will be explored. Thus, science is also progressive and tentative. Science is concerned with theory, and is parsimonious.
question
What is the principle of parsimony in science?
answer
Parsimony means stinginess. This principle holds that we should use the simplest explanation possible to account for a given phenomenon.
question
Briefly describe the working assumptions of science.
answer
Realism, rationality, regularity, discoverability, and causality. Scientists assume: (1) that the world has an existence outside our mind; (2) that the world can be understood by logical reasoning; (3) that the world follows the same laws at all times and at all places; (4) that we can discover how the world works; and (5) that every event has a cause.
question
Briefly describe the three processes in the discovery of regularities.
answer
Description: We must define the subject matter; events and entities such as s-r, cognitions, etc. Discovery of lawful relationships: Can be absolute or probabilistic, causal or correlational. Search for causes: The most important process, the end goal.
question
What is a law? Give an example of a law.
answer
A law is a statement that certain events are regularly associated with each other in an orderly way.
question
Define a theory, both in broad terms and in strict terms.
answer
Broadly, a sceintific thoery is a statement or set of statements about the relationship between variables. If the statements concern only a single relationship between variables, we are speaking of a law. However, sometimes a number of laws are tied together into a more general set of statements, which is called a theory. The stricter sense is that theory is a statement or set of statements about relationships among variables that includes at least one concept that is not directly observed but that is necessary to explain these relationships.
question
According to Sir Karl Popper, can a theory be proven true? Why, or why not?
answer
According to Sir Karl Popper, a scientific theory can never be proven true because many other theories can predict any given outcome. The most fruitful test of a theory is to set up a condition in which it can be disconfirmed. If the theory survives the test, we can gain more confidence in it
question
What roles do theories play in science?
answer
Theories play three major roles in science. The first role is to organize existing knowledge and explain laws. The second role is to predict new laws. The third role is to guide research. In other words, a constant interplay exists between theory and empirical observation. A theory is originally based on empirical observations, and, once it has been developed, predictions about the theory must be tested through empirical observations. If the predictions are confirmed, evidence is provided that the theory is useful as an explanation for the given event. If the predictions are refuted, either the theory has to be abandoned or modified to account for the data, or a more sensitive test of the theory needs to be devised.
question
What is a scientific hypothesis? How is a hypothesis related to a theory? Give an example.
answer
A hypothesis is a prediction from theory and said to be true for the purpose of testing its validity. A hypothesis can be put in the form of an if-then statement: If A is true, than B should follow. Hypotheses test predictions made from theories. For example, if the oedipus complex is true (theory) than we should find desire for one's opposite sex parent ubiquitous among children.
question
What is an operational definition? Why is it desirable to use convergent operations?
answer
An operational definition is a statement of the precise meaning of a procedure or concept within an experiment. Using different ways of defining a concept via different operational definitions is called converging operations. Using converging operations is desirable because with each addition of the concept less objections can be made about its limitations, until a high degree of confidence in the operational definition is reached.
question
What is a paradigm?
answer
A paradigm is a set of laws, theories, methods, and applications that form a scientific research tradition; for example, Pavlovian conditioning.
question
List and explain the purpose and procedures in reviewing the literature.
answer
Literature reviews tell one what is already known, to ensure that science is actually cumulative. Check for books of interest in the library catalogue, consult textbooks and handbooks, and search journals. Skim through them until you have a general idea of what is known on your topic, narrow your interest, and then dig deeper to determine what work has been done on the specific question you would like to answer.
question
How are variables related to theoretical concepts? Give an example.
answer
On the empirical level, a variable is related to an abstract theoretical concept by means of the operational definitions used to measure the concepts. Thus, anxiety may be defined as galvanic skin response (GSR) readings, which comprise a variable that can be measured. Defining a theoretical concept by one operation alone can be limiting, however, and multiple operational definitions are often necessary to capture the full meaning of the concept.
question
Distinguish between independent variables and dependent variables. Give examples.
answer
In an experiment, the independent variable is one that the experimenter changes deliberately. The values of an independent variable define its treatment conditions. The dependent variable is the variable that measures the impact of changes in the independent variable. A confounding variable is one that varies systematically with the independent variable (Christensen, 2007). When such variables are present, it is impossible to determine whether any observed differences among the different treatment conditions are caused by changes in the independent variable, or by the influence of the confounding variables, or by both. A manipulated variable is an independent variable that is under the direct control of the experimenter. A subject variable is one that reflects measurable subject characteristics, such as age and sex. When researchers study the relationships between subject variables and behaviour, they are doing correlational research, not experimental research.
question
What are the levels of an independent variable?
answer
The degree to which an independent variable is present or absent. For example, whether one is exposed to high, medium, low, or no television violence.
question
What are the considerations when choosing the levels of an independent variable?
answer
Considerations would be what is feasible, safe, and measurable.
question
Distinguish between quantitative and categorical variables. Give examples.
answer
A quantitative variable varies in amount (how much food), whereas a categorical variable varies in kind (what type of food).
question
What is the difference between continuous and discrete variables?
answer
A continuous variable is one that falls along a continuum and is not limited to a certain number of values (time). A discrete variable is one that falls into seperate bins with no intermediate values possible (day of the week).
question
Define measurement.
answer
Measurement is essentially the assignment of numbers to events or objects according to some rules.
question
Describe the four types of measurement scales. What are their properties? Give examples.
answer
Measurement scales have one or more of the following properties: difference or identity, magnitude, equal intervals, and true zero. Stevens classified variables into four measurement levels or scales: the nominal, ordinal, interval, and ratio scales. The nominal scale, which is the lowest level, serves a classification function only (vegetables). The ordinal scale possesses the magnitude property in addition to the difference property, and ranks objects or events in order of the magnitude of an attribute (ranking and type). For the interval scale, equal numerical intervals reflect equal magnitudes of the variable being measured. It has the properties of difference, magnitude, and equal intervals (differences between numbers are meaningful). At the highest level is the ratio scale, which has all four properties. The ratio scale has a true zero point, meaning that zero represents the true absence of what is being measured. As McBurney and White (2010) point out, psychological variables are measured at nominal, ordinal, and, at best, interval levels.
question
Is a person with an IQ of 120 \"twice as smart\" as a person with an IQ of 60? Why, or why not?
answer
IQ is an interval scale. IQ tests are designed so that the amount of the differences between people can be meaningfully represented by the IQ score. However, because IQ is not a ratio scale it would be meaningless to say that a 120 = twice as smart as 60.
question
Define reliability of a test or measure. What are the types of reliability of measures? Give an example of each.
answer
A reliable measurement is one that shows consistency over time. Inter-rater reliability assesses the degree to which different judges agree in their assessment decisions. This is useful because human observers will not necessarily interpret answers the same way. Parallel forms reliability is a measure of reliability obtained by administering different versions of an assessment tool (both versions must contain items that probe the same construct, skill, knowledge base, etc.) to the same group of individuals. The scores from the two versions can then be correlated in order to evaluate the consistency of results across alternate versions. Test-retest reliability is a measure of reliability obtained by administering the same test twice over a period of time to a group of individuals. The scores from Time 1 and Time 2 can then be correlated in order to evaluate the test for stability over time. Internal consistency reliability is a measure of reliability used to evaluate the degree to which different test items that probe the same construct produce similar results.
question
Define validity of a measurement. Give an example of each type of validity in relation to tests and measurement.
answer
A valid measurement is one that reflects accurately the variable under study. Construct validity is the degree to which a test actually measures the constructs it was designed to measure, but not others. Face avlidity is the idea that a test should appear to any person to be a test of what it is supposed to test. Content validity is the notion that a test should sample the range of the behavior that is represented by the theoretical concept being measured. An intelligence test, for eample, should measure general knowledge, verbal ability, spatial ability, and quantitative skills, among others. An intelligence test that measured only spatial ability would not have sufficient content validity. Criterion validity is the idea that a valid test should relate closely to other measures of the same theoretical construct. A valid test of intelligence should correlate highly with other intelligence tests. It should also correlate with behaviors that are considered to require intelligence, such as doing well in school. Note that a measurement can be reliable but invalid, but a measurement cannot be both unreliable and valid. Both reliability and validity of a measurement can be drastically reduced if a measurement is reactive; that is, if the act of measuring influences the behaviour or thought that one is measuring.
question
Define internal validity.
answer
Internal validity refers to the degree to which the experimental manipulation alone caused the change in behaviour.
question
What is meant by confounding?
answer
Confounding is an error that occurs when the effects of two variables in an experiment cannot be seperated, resulting in a confused interpretation of the results
question
Why is confounding particularly acute in research in which a subject variable is used?
answer
This is because researchers cannot randomly assign subject variables such as gender or race, thus confounding is highly likely due to differential treatment, exposure to media, social class, etc.
question
Define construct validity.
answer
Construct validity refers to the degree to which the study measures and manipulates the underlying concepts the researcher claims. One must rule out other possible theoretical explanations of results
question
Briefly describe the several ways to determine if a test has construct validity.
answer
1. Manipulation check, which aims to see that a viariable is working in the way you think it is i.e., using standardized tests for measuring the variable as well as your own 2. Use various measurements of the construct to determine if it, and not another related variable is being altered (i.e., is the rat hungry or not satiated?)
question
Define external validity.
answer
External validity refers to the degree to which the results could be generalized to different subjects, settings, or times.
question
Define statistical validity.
answer
Statistical validity refers to the degree to which the observed results are not due to chance only.
question
Briefly describe the major threats to internal validity.
answer
Internal validity refers to the extent to which a study provides evidence of a cause and effect relationship between the indpendent and dependent variables. 1. Events outside the lab: Time of day, day of the week, national tragedies, etc. 2. Maturation: People get older and behave differently 3. Effects of testing: Learn how to take tests, become less anxious, etc. 4. Regression effect: Regression towards the mean, those with extreme scores tend to drift towards the center 5. Selection: those chosen for one condition may be different in an important way from those who were placed in the other condition 6. Mortality: participants dropping out of the study may be different in some way than those who complete it.
question
Briefly describe two threats to construct validity.
answer
Construct validity is the extent to which the results support the theory behind the research. This is the most difficult type of validity to achieve because of the indefinite number of theories that may account for a given lawful relationship. Loose connection between theory and method: i.e., using poor measures of theoretical constructs. Ambiguous effect of independent variable: participant expectations can alter their responding, people have a good-subject tendency, and often have evaluation apprehension. The general strategy for obtaining construct validity in a piece of research is to ask whether alternative theoretical explanations of the data are less plausible than the theory believed to be supported by the research.
question
Briefly describe three threats to external validity.
answer
1. Other subjects: Findings may not apply to everyone as people are different in many important respects and some groups are overstudied while others are understudied 2. Other times: society changes over time, and thus people's reactions to certain things will change 3. Other settings: Lab work may not transpose to natural settings and vice versa.
question
What are the two broad categories of bias resulting from the interaction between subject and experimenter?
answer
Role demands: participant's expectations of what an experiment requires them to do. Experimenter bias: when an experimenter's wishes or expectations influence the results of a study
question
What are role demands ( or demand characteristics)? How might they be overcome?
answer
The act of being in a study will influence people to do things they would not otherwise, and to not do things they would otherwise. Keeping the real aim of the experiment secret or to use a measure that is unlikely to be influenced by participant's guesses about the hypothesis.
question
How might experimenter bias be overcome?
answer
Double blinding and standardizing the experiments as much as possible will reduce experimenter bias.
question
What does the concept of control mean in experimental research?
answer
Control is any means used to rule out possible threats to the validity of a piece of research. In psychology, the concept of control is used in two different ways, first as a standard of comparison, and second, as a way of reducing variability. Essentially, it is a way of establishing that two individuals are identical except for the variable(s) of interest.
question
Distinguish the meanings of control: control experiment and experimental control.
answer
The primary meaning allows one to conclude that a dependent variable is associated with an independent variable and not any other variable (control experiment). THe second usage facilitates drawing this conclusion by so limiting the number of variables operating in the situation and their range of calues that the conclusion is clearer (experimental control).
question
List several specific strategies for achieving control.
answer
The most important and basic of the control strategies is randomization. This is a statistical control technique designed to ensure that extraneous variables, known or unknown, will not systematically bias the results. Another strategy is matching of subjects in the various experimental treatment groups. An advantage of matching is that it can increase the sensitivity of an experiment; that is, the ability to detect the influence of the independent variable regardless of how small its effect may be. A third strategy involves building the extraneous variable into the research design to allow any effects of the extraneous variable in question to be examined. Finally, statistical adjustments can be made with strategies or techniques such as analysis of covariance to equate subjects on paper.
question
Briefly describe the strategy of subject as own control. What are the limitations of this strategy?
answer
Each subject acts as her or his own control group, thus eliminating the confound of differences between subjects. This is not possible when contrast effects exist such that experiencing one condition will influence the response to the other condition, when one is comparing two different methods of doing something (i.e., learning group of vocabulary words with and without mnemonics), or when one needs to do a statistical analysis.
question
What is random assignment? How can random assignment be performed?
answer
Random assignment is unbiased assignment process that gives each subject an equal and independent chance of being placed in every condition. You can give each participant a number, then look at a random number table to determine which participants will go in which category.
question
Why is matching a strategy for achieving control? Under what conditions should matching of subjects be done? Describe the proper matching procedure.
answer
Matching involves ensuring that experimental and control groups are equated on one or more variables before the experiment, generally those that are strongly correlated with the dependent variable. This is useful when subjects differ among themselves on an independent variable known or suspected to effect the dependent variable of interest. Second, it must be feasible to test participants on this variable. Third, it is possible to weaken your experiment by matching the subjects if the matching variable is not substantially correlated with the dependent variable. This effects results because the statistical test appropriate for a matched-groups design considers the data from pairs of subjects, whereas the randomized groups test considers individual subjects. You can see that there are twice as many subjects as pairs of subjects, so the randomized test has more numbers to work with and is therefore more powerful. You determine the levels of the variable by which participants will be matched, make pairs of the closest clusters of two, and randomly assign members of each pair to the control and experimental groups.
question
Briefly describe building nuisance variables in an experiment as a strategy for achieving control.
answer
A nuisance variable is a condition that cannot be easily removed and so is made an independent variable as a means of control. This could involve seperating the two groups and comparing their interaction with the original independent variable seperately.
question
How is statistical control used as a strategy for achieving control?
answer
Statistical control is a matehmatical means of comparing subjects on paper when they cannot be equated as they exist in fact. This involves determining if the experiment is statistically valid and what conclusions can be drawn from further statistical analysis.
question
What is replication? Distinguish two types of replication: direct and systematic.
answer
Replication is the repeating of an experiment to see if the results are the same. Direct replication involves following the original research design and methodology, whereas systematic replication will expand on the first study by testing something that should happen if the first one's results were valid.
