Chapter 5
How can you correct problems associated with convenience sampling?
Try to ensure that the samples are reasonably representative and not strongly biased Provide a clear description of how the sample was obtained and who the participants are Let everyone make their own judgments about representativeness
What is the Little Genius Problem?
When studying children, you may need to recruit parents whoare willing to participate in research Typically, they know a lot about research because they themselves are highly educated The issue: While our clinical samples may be representative, our comparative samples are typically of higher SES, have parents with higher education, and thus may look less like the population you're trying to generalize to
The process of selecting individuals for a study
Sampling
What are the two principle METHODS of random sampling?
Sampling with replacement Sampling without replacement
However, it is more likely that a biased sample is the result of ____, which means that the sampling procedure favors the selection of some individuals over others
Selection bias E.g., if we recruit from a university, we are likely to obtain a sample that is smarter, on average, than individuals in the entire population
The selection process is fair and unbiased, but there is no guarantee that the sample is representative.
Simple random A sample is obtained using a random process to select participants from a list containing the total population. The random process ensures that each individual has an equal and independent chance of selection.
What are the 5 probability sampling methods?
Simple random sampling Systematic sampling Stratified random sampling Proportion stratified random sampling Cluster sampling Combined-strategy
The accuracy of a sample's representation increases in relation to the ____ of the sample size
Square root
Advantage of proportionate stratified random sampling?
*The sample can be constructed so that several variables (e.g., age, economic status, political affiliation, etc.) are represented in the sample in the same proportions in which they exist in the population
What are the negatives of this method?
A lot of extra work -Requires a lot of preliminary measurement before the study actually begins Discards many of the sampled individuals Can make it impossible for a researcher to describe or compare some subgroups or strata that exist within the population E.g., if a specific subgroup makes up only 1% of the population, they must only make up 1% of the sample. If there's 100 individuals in the sample, this means that there is only 1 person from the subgroup.
Dr. Near conducts an experiment on memory for individuals who are above the age of 65. Although there are millions of people above the age of 65, she selects a group of 25 to participate in the experiment. What name is given to the group of 25?
A sample
Guarantees that each subgroup will have adequate representation, but the overall sample is usually not representative of the population
A sample is obtained by dividing the population into subgroups (strata) and then randomly selecting equal numbers from each of the subgroups
What name is given to the group of individuals from which researchers actually select participants for research studies?
Accessible population
A portion of the target population who are accessible to be recruited
Accessible population E.g., adolescents diagnosed with bulimia nervosa at local clinics and agencies
Researchers often begin a research study with a specific target for the level of ____
Accuracy E.g., a political poll may want a margin of error of 5% in determining voters' preferences between two candidates In this situation, it can be computed that the sample must have at least 384 individuals to be confident that the preferences observed in the sample are within 5% of the corresponding population preferences
Why is quota sampling NOT the same as stratified and proportionate stratified sampling?
Because it does not randomly select individuals from the population
The major threat to select a representative sample is ____
Bias
A ____ is one that has characteristics noticeably different from those of the population
Biased sample E.g., The individuals in a sample are smarter, older, faster... than the individuals in the population
How is the n calculated in systematic sampling?
By dividing the population size by the desired sample size E.g., population of 100 children, desired sample size of 25... size of n is 100/25 = 4
A researcher recruits a sample of 25 preschool children for a research study by posting an announcement in a local daycare center describing the study and offering a $10 payment for participation. What kind of sampling is the researcher using?
Convenience sampling
Most commonly used sampling method in behavioral science research It's easy Inexpensive
Convenience sampling
Researchers use participants those individuals who are easy to get People are selected on the basis of their availability and willingness to respond
Convenience sampling
A biased sample can occur simply by ____
Chance
What are the steps of random sampling?
Clearly define the population from which you want to select a sample List all the members of the population Use a random process to select individuals from the list (e.g., random number table)
An easy method for obtaining a large, relatively random sample, but the selections are not really random or independent.
Cluster Instead of selecting individuals, a sample is obtained by randomly selecting clusters (preexisting groups) from a list of all the clusters that exist within the population.
A probability sampling technique involving random selection of GROUPS instead of individuals
Cluster sampling E.g., A researcher may want to obtain a large sample of 3rd grade students from the city school system. Instead of selecting 300 students at 1 time, the researcher can randomly select 10 classrooms (each with about 30 students) and still end up with 300 individuals in the sample
A teacher obtains a sample of children from a fifth-grade classroom by randomly selecting the third, fifth, and eighth rows and taking all of the students in those rows. What kind of sampling is being used?
Clustered sampling
What is the college sophomore problem?
College students in an introduction to psychology course have to participate in studies to get credit; therefore, we have the SAME students in dozens of our tasks Mainly an issue when you have typical adults do tasks to learn more about them or specific constructs, not when you're doing research on clinical populations The issue: college students may not be very representative
What are some potential solutions of the college sophomore problem?
Consider the makeup of your convenience sample Try to recruit from a diverse array of students Acknowledge the limitation and move on
An easy method for obtaining a sample, but the sample is probably biased.
Convenience A sample is obtained by selecting individual participants who are easy to get.
Which of the following sample techniques is most likely to result in a biased sample? Simple random sampling Convenience sampling Proportionate stratified random sampling Systematic sampling
Convenience sampling
What are the 2 non-probability sampling methods?
Convenience sampling Quota sampling
The average difference between the sample mean and the population mean ____ as the sample size ____ However, there is a limited benefit from increasing sample size beyond a certain threshold of participants (usually beyond 25-30)
Decreases Increases
Steps of proportionate stratified random sampling?
Determine the desired size of the sample Randomly select from a subgroup in the population until you have a number corresponding to the percentage of that subgroup (e.g., freshman are 30% of college population, you want 10 participants in your sample, randomly select 3 freshman) Continue this process with each subgroup to obtain the full sample
Social media can be a useful tool to get people to participate in online surveys But BE CAUTIOUS Certain tools are better than others
Do not provide public links Screen participants and then email those who passed the screener with a unique survey link
Why is convenience sampling considered a WEAK form of sampling?
Does NOT require knowledge of the population Does NOT use a random process for selection Strong possibility that the sample is BIASED
What is the basic requirement for random sampling?
Each individual in the population has an equal chance of being selected *Equality
What is the second requirement for random sampling?
Each selection is independent of the others I.e., the choice of one individual does not influence or bias the probability of choosing another individual *Independence
Usually, target populations are not ____
Easily available (i.e., do not have access to every adolescent in the world with bulimia nervosa)
What is the goal of random sampling?
Ensure that the selection procedure cannot discriminate among individuals and thereby result in a non-representative sample
Another factor influence sample size in research is ____
Ethics If a sample is too large, then it is unnecessarily using extra subjects or participants, which can be viewed as unethical
Be cautious about ____ results of a study using the accessible population to the target population
Generalizing
What is the main advantage of stratified random sampling?
Guarantees that each of the different subgroups will be well represented with a relatively large group of individuals in the sample Thus, this type of sampling is appropriate when the purpose of the research study is to examine specific subgroups and make comparisons between them
To share/distribute flyers via social media, it must be approved in your ____ application
IRB
Steps of stratified random sampling
Identify the specific subgroups (or strata) to be included in the sample Select equal-sized random samples from each of the pre-identified subgroups, using the same steps as in simple random sampling Combine the subgroup samples into one overall sample
Because chance determines each selection, it is possible (although usually unlikely) to obtain a very distorted sample E.g., we could get a random sample of 10 males from a population of equal number of men and women What do researchers do to avoid this?
Impose additional restrictions on the random sampling procedure
Disadvantages of cluster sampling?
It can raise concern about the independence of individual scores Individuals within a cluster often have common characteristics or share common experiences that may influence the variables being measured
A ____ sample is probably more representative than a ____ sample
Large Small
____: states that the larger the sample size, the more likely it is that values obtained from the sample are similar to the actual values for the population
Law of large numbers
How large should the sample be to be representative?
No simple answer BUT there are some general guidelines to help choose a sample size
____ sampling: the odds of selecting a particular individual are UNKNOWN because the researcher does not know the population size and cannot list the members of the population
Non-probability
For situations in which the researcher cannot know the complete list of potential participants, what kind of sampling is necessary?
Non-probability sampling
Individuals in a target population typically share ____
One characteristic E.g., children of divorced parents, adolescents with bulimia nervosa
Bottom line: not everyone can participate, and the ____ of the study may depend on the way in which participants are selected
Outcome
Example of stratified random sampling?
Plan to select 50 individuals from a large introductory psychology class and want to ensure that psychology majors and nonmajors are equally represented Select a random sample of 25 students from the psychology majors in the class Select a random sample of 25 students who are not psychology majors Combine the two subgroup samples to produce the desired stratified random sample
What is proportionate stratified random sampling commonly used for?
Political polls, major public opinion surveys, etc.
The large group of interest
Population Typically, populations are huge (e.g., adolescents, men, women, or humans)
Samples tend to be similar to the ____ from which they are taken
Populations E.g., If you select a sample population for which the average age is 21 years, you will probably get a sample with an average age around 21 years
Small set of individuals who participate
Sample
____ sampling: the odds of selecting a particular individual are known and can be calculated
Probability E.g., probability selection of 1/100 for each person
What are the two basic categories of sampling methods?
Probability Non-probability
Guarantees that the composition of the sample (in terms of the identified strata) will be perfectly representative of the composition of the population, but some strata may have limited representation in the sample.
Proportionate stratified A sample is obtained by subdividing the population into strata and then randomly selecting from each stratum a number of participants so that the proportions in the sample correspond to the proportions in the population.
A probability sampling technique that involves identifying specific subgroups to be included, determining what PROPORTION of the population corresponds to each subgroup, and randomly selecting individuals so that the proportion for each subgroup in the sample exactly matches the corresponding portion in the population
Proportionate stratified random sampling E.g.., College administration wants a sample of students that accurately represent the distribution of freshmen, sophomores, juniors, and seniors in the college population Sample is then selected so that it has exactly the same percentages for the 4 groups
A variation of quota sampling mimics ____
Proportionate stratified sampling
Allows a researcher to control the composition of a convenience sample, but the sample probably is biased.
Quota A sample is obtained by identifying subgroups to be included, then establishing quotas for individuals to be selected through convenience from each subgroup.
A non-probability sampling method; a type of convenience sampling involving identifying specific subgroups to be included in the sample and then establishing quotas for individuals to be sampled from each group E.g., adjust the quotas to ensure that the sample proportions match a predetermined set of population proportions
Quota sampling
A researcher would like to select a sample of 50 people so that five different age groups are equally represented in the sample. Assuming the researcher DOES NOT KNOW the entire list of people in the population, which sampling technique should be used?
Quota sampling
Selection must be a ____ process, which ensures that every possible outcome is equally likely
Random
If each person in a large group has an equal chance of being included in an experiment, then what kind of sampling is being used?
Random sampling
The goal of a research study is to examine the ____, then generalize the results to the entire ____
Sample Population
Advantages of cluster sampling?
Relatively quick and easy way to obtain a large sample The measurement of individuals can be done in groups, which can greatly facilitate the entire research project
*Most populations are so large than the probabilities remain unchanged from one selection to the next, even when we don't ____ individuals
Replace
Although samples tend to be similar to their populations, most researchers would like more assurance that the samples in their research study are good representatives of the populations they want to study to increase the likelihood of obtaining a ____
Representative outcome
To generalize the results of a study to a population, the researcher must select a ____
Representative sample
How accurately we can generalize the results from a given sample to the population depends on the ____ of the sample
Representativeness This refers to how closely the sample mirrors or resembles the population
If possible, researchers should attempt to perform some preliminary statistical calculations to determine whether the expected results are likely to be ____
Statistically significant
Step ___: Identify the participants or subjects for the study, decide how they will be selected, and plan for their ethical treatment
Step four
A probability sampling technique that involves identifying specific subgroups to be included in the sample and then selecting equal-sized random samples for each pre-identified subgroup
Stratified random sampling
A researcher would like to describe and compare the attitudes of four different ethnic groups of students at a local state college. What kind of sampling would be best to obtain the participants for this study?
Stratified random sampling
An easy method for obtaining an essentially random sample, but the selections are not really random or independent.
Systematic A sample is obtained by selecting every nth participant from a list containing the total population after a random start.
A probability sampling technique in which a sample is obtained by selecting every nth participant from a list containing the total population after a random starting point
Systematic sampling
The group defined by the researcher's specific interests
Target population
What are the negatives of stratified random sampling?
Tends to produce a distorted picture of the overall population -Subgroups may not be equally represented in the population Every individual in the population does NOT have an equal chance of being selected for the sample -Not a true random sample because all individuals in the population are not equally likely to be selected
What are the 3 important conditions of probability sampling?
The exact size of the population must be known and it must be possible to list all of the individuals Each individual in the population must have a specified probability of selection When a group of individuals are all assigned the same probability, the selection process must be unbiased so that all group members have an equal chance of being selected.
Why is systematic sampling less random than simple random sampling?
The principle of INDEPENDENCE is violated BUT ensures a high degree of representativeness E.g., if we select participant 11, we are biased in favor of choosing participant 15
What are some potential solutions of the little genius problem?
Try to recruit elsewhere instead of professors/students with kids Use Vanderbilt resources (with caution) Do community outreach Acknowledge limitation and move on
In non-probability sampling, the researcher does not use an ____ method of selection
Unbiased Thus, non probability sampling has a greater risk of producing a biased sample than does probability sampling
What is the Professional Participant Problem?
We often get the same people over and over again because we have the same resources You MIGHT get a more representative sample (lots of families find out about research, so it may not be as biased) but they are better at certain tasks than others due to repetition and may be familiar with your methods
When is stratified random sampling useful?
When a researcher wants to describe each individual segment of the population or wants to compare segments
Sampling _____ replacement: Individual selected is recorded as a sample member and then returned to the population Ensures that the probability of selection remains constant/independent selections (e.g., 1/100)
With
Sampling ____ replacement Individual selected is removed from the population before the next selection is made Ensures no individual appears more than once in a single sample However, the probability of selection changes (e.g., 1/100 → 1/99)
Without