Lesson Transcript
Instructor: Devin Kowalczyk
Devin has taught Psychology and has a master's degree in Clinical Forensic Psychology, and will earn a PhD in 2015.
When planning an experiment, you will likely use groups of participants. This lesson explores the types of groups an experimenter can collect data from and the reason why there are different groups.
Participants
When you conduct an experiment or survey you collect information from a group of people. Now, while 'group of people' may seem like an adequate description, it is, in fact, not. We need more a specific term because the statistics we use are different depending on group we use. But don't worry, there's no complicated process to identifying the group of people you use.
The first group of people is a population, which is defined as the complete collection to be studied. The second group is a sample, which is defined as a section of the population. Let's look at some examples to help make this a little clearer.
Population
When you are attempting to study a population, you have to collect information from everyone in that group. This makes it extremely difficult to study populations. For example, imagine if you are going to study:
All people with schizophrenia in the United States (approximately 3.1 million individuals).
Californians' view on raisins (38 million).
Immigrants beliefs about the U.S.'s foreign policy (nobody really knows how many if you include illegal and legal).
So you can see the difficulty with studying populations. Smaller populations, like community colleges, are easier to collect population data. However, the issue here is if you have one person who fails to contribute, then you don't have a population.
Sample
Most social researchers realize that obtaining information from every person in a population is next to impossible. So instead of trying to collect everyone's information, they collect a sample of the population. But unlike a population, which is everyone, there are different ways you can collect a sample of a population. The different ways of taking a sample are sort of like how there are different ways to cut a cake. Here is a list of the different sampling techniques:
Random sample: each individual in the population has an equal chance of being selected.
Stratified sample: a researcher divides the population into groups based on characteristics, and then the researcher randomly selects from each group based on its size.
Quota sample: a researcher deliberately sets a requirement to ensure a particular group is represented.
Purposive sample: a researcher purposefully focuses on a particular subset of a population.
Convenience sample: selection of the sample is based on ease of accessibility.
So, enough talking about cake; what would these samples look like if they were real scientific studies? Let's say you're interested in studying people with schizophrenia in the United States.
A random sample would mean that each person with schizophrenia has an equal chance of being part of your study. This might mean a lot of travel since the U.S. is a big place.
A stratified sample would be useful if you were interested in looking at a particular subset of people with schizophrenia. What if you were interested in comparing the level of symptoms between upper socioeconomic status and lower socioeconomic status?
Instructor: Devin Kowalczyk
Devin has taught Psychology and has a master's degree in Clinical Forensic Psychology, and will earn a PhD in 2015.
When planning an experiment, you will likely use groups of participants. This lesson explores the types of groups an experimenter can collect data from and the reason why there are different groups.
Participants
When you conduct an experiment or survey you collect information from a group of people. Now, while 'group of people' may seem like an adequate description, it is, in fact, not. We need more a specific term because the statistics we use are different depending on group we use. But don't worry, there's no complicated process to identifying the group of people you use.
The first group of people is a population, which is defined as the complete collection to be studied. The second group is a sample, which is defined as a section of the population. Let's look at some examples to help make this a little clearer.
Population
When you are attempting to study a population, you have to collect information from everyone in that group. This makes it extremely difficult to study populations. For example, imagine if you are going to study:
All people with schizophrenia in the United States (approximately 3.1 million individuals).
Californians' view on raisins (38 million).
Immigrants beliefs about the U.S.'s foreign policy (nobody really knows how many if you include illegal and legal).
So you can see the difficulty with studying populations. Smaller populations, like community colleges, are easier to collect population data. However, the issue here is if you have one person who fails to contribute, then you don't have a population.
Sample
Most social researchers realize that obtaining information from every person in a population is next to impossible. So instead of trying to collect everyone's information, they collect a sample of the population. But unlike a population, which is everyone, there are different ways you can collect a sample of a population. The different ways of taking a sample are sort of like how there are different ways to cut a cake. Here is a list of the different sampling techniques:
Random sample: each individual in the population has an equal chance of being selected.
Stratified sample: a researcher divides the population into groups based on characteristics, and then the researcher randomly selects from each group based on its size.
Quota sample: a researcher deliberately sets a requirement to ensure a particular group is represented.
Purposive sample: a researcher purposefully focuses on a particular subset of a population.
Convenience sample: selection of the sample is based on ease of accessibility.
So, enough talking about cake; what would these samples look like if they were real scientific studies? Let's say you're interested in studying people with schizophrenia in the United States.
A random sample would mean that each person with schizophrenia has an equal chance of being part of your study. This might mean a lot of travel since the U.S. is a big place.
A stratified sample would be useful if you were interested in looking at a particular subset of people with schizophrenia. What if you were interested in comparing the level of symptoms between upper socioeconomic status and lower socioeconomic status?
Category
📚
Learning