Hi Hemangi sorry i m posting it late, your question was for the genralisibility and representativeness of a sample which type of stratified sampling would we choose?
Any type of stratified sampling can be choosen because we select sample randomly so the sample will be representative for the population and results can be genralised but proportional stratified sampling will be more appropriate because there will be more representative samples because samples will be selected proportioanate to the population and it also reduces errors between strata with respect to the relative numbers selected.
Sunday, April 29, 2007
Thursday, April 19, 2007
Stratified sampling
Stratified sampling
Since the word stratify means arranging in a sequence of grades or ranks. So we think that the stratified sampling can be the selection of samples on the basis of certain criteria when the population is randomly selected. Criteria for selection can be anything like their characteristics, physical location, socio economic status, etc.
Meaning:-
The population is divided into strata and a random sample is taken from each stratum.
Stratification means division of the universe into groups according to geographical, sociological, or economic characteristics. When the population is heterogeneous with respect to the variable or characteristics under study then the technique of stratified sampling is use to obtain more efficient and accurate results.
Stratified sampling is where we begin by grouping elements that share certain characteristics, or dividing the population into several large groups, or clusters. Its purpose is to classify populations into subpopulations or strata based on some supplementary information and then a selection of separate samples from each of the strata.
The two types of stratified sampling are
1. Proportionate stratified
2. Disproportionate stratified
1. Proportional Stratified:
In this type, the strata sample size is made proportional to the strata population size. It is used to get a more representative sample than might be expected under Simple Random Sampling. Reduces sampling errors between strata with respect to the relative numbers selected. This is true when there are homogeneous groups. Population strata must be known in order to draw a proportionate stratified sample.
2. Disproportionate stratified:
It is used to manipulate the number of cases selected in order to improve efficiency of the design. The main interest is to study separate sub-populations represented by the strata rather than on the entire population. Here a varying sampling is used. In disproportional stratified sampling, the sub samples are not proportional to their sizes in the population.
Here is an example showing the difference between proportional and disproportional stratified sampling:
If the population is 75% female and 25% male. And if a researcher wants a sample of size 100 and wants to stratify on the variable called gender.
For proportional stratified sampling, researcher would randomly select 75 females and 25 males from the population.
For disproportional stratified sampling, researcher might randomly select 50 females and 50 males from the population.
Advantages:
Stratification will always achieve greater precision provided that the strata have been chosen so that members of the same stratum are as similar as possible in respect of the characteristic of interest. The bigger the differences between the strata, the greater the gain in precision. For example, if you were interested in Internet usage you might stratify by age, whereas if you were interested in smoking you might stratify by gender or social class.
It is often administratively convenient to stratify a sample. Interviewers can be specifically trained to deal with a particular age-group or ethnic group, or employees in a particular industry. The results from each stratum may be of intrinsic interest and can be analyzed separately.
It ensures better coverage of the population than simple random sampling.
Disadvantages:
Difficulty in identifying appropriate strata.
More complex to organize and analyze results.
References:
http://www.southalabama.edu/coe/bset/johnson/lectures/lec7.htm
http://www.mis.coventry.ac.uk/~nhunt/meths/strati.html
From,
Vipula Khanolkar 07
Aditee Moghe 11
Since the word stratify means arranging in a sequence of grades or ranks. So we think that the stratified sampling can be the selection of samples on the basis of certain criteria when the population is randomly selected. Criteria for selection can be anything like their characteristics, physical location, socio economic status, etc.
Meaning:-
The population is divided into strata and a random sample is taken from each stratum.
Stratification means division of the universe into groups according to geographical, sociological, or economic characteristics. When the population is heterogeneous with respect to the variable or characteristics under study then the technique of stratified sampling is use to obtain more efficient and accurate results.
Stratified sampling is where we begin by grouping elements that share certain characteristics, or dividing the population into several large groups, or clusters. Its purpose is to classify populations into subpopulations or strata based on some supplementary information and then a selection of separate samples from each of the strata.
The two types of stratified sampling are
1. Proportionate stratified
2. Disproportionate stratified
1. Proportional Stratified:
In this type, the strata sample size is made proportional to the strata population size. It is used to get a more representative sample than might be expected under Simple Random Sampling. Reduces sampling errors between strata with respect to the relative numbers selected. This is true when there are homogeneous groups. Population strata must be known in order to draw a proportionate stratified sample.
2. Disproportionate stratified:
It is used to manipulate the number of cases selected in order to improve efficiency of the design. The main interest is to study separate sub-populations represented by the strata rather than on the entire population. Here a varying sampling is used. In disproportional stratified sampling, the sub samples are not proportional to their sizes in the population.
Here is an example showing the difference between proportional and disproportional stratified sampling:
If the population is 75% female and 25% male. And if a researcher wants a sample of size 100 and wants to stratify on the variable called gender.
For proportional stratified sampling, researcher would randomly select 75 females and 25 males from the population.
For disproportional stratified sampling, researcher might randomly select 50 females and 50 males from the population.
Advantages:
Stratification will always achieve greater precision provided that the strata have been chosen so that members of the same stratum are as similar as possible in respect of the characteristic of interest. The bigger the differences between the strata, the greater the gain in precision. For example, if you were interested in Internet usage you might stratify by age, whereas if you were interested in smoking you might stratify by gender or social class.
It is often administratively convenient to stratify a sample. Interviewers can be specifically trained to deal with a particular age-group or ethnic group, or employees in a particular industry. The results from each stratum may be of intrinsic interest and can be analyzed separately.
It ensures better coverage of the population than simple random sampling.
Disadvantages:
Difficulty in identifying appropriate strata.
More complex to organize and analyze results.
References:
http://www.southalabama.edu/coe/bset/johnson/lectures/lec7.htm
http://www.mis.coventry.ac.uk/~nhunt/meths/strati.html
From,
Vipula Khanolkar 07
Aditee Moghe 11
Saturday, April 14, 2007
Difference between examples
Difference between first example and second example
First example:
· In this example, researcher decides to carry out a survey.
· In this example, the sample is selected randomly that is out of 1000 primary school
teachers in Mumbai 300 are randomly selected.
· In this there is only one group that is 300 randomly selected samples on basis of which her
survey will be done.
Second example:
· In this example, researcher decides to carry out a study that is the actual experiment is
carried out by giving treatment to the group.
· In this example, the selection of sample is random but with different method the
sample is selected that is researcher lists everyone’s names on chits, and draws one chit
out of the lot and writes that name under a column named linear CAI and puts that chit
aside.
· In this, there are two groups since there are two different CAI packages of which
effectiveness has to be checked
First example:
· In this example, researcher decides to carry out a survey.
· In this example, the sample is selected randomly that is out of 1000 primary school
teachers in Mumbai 300 are randomly selected.
· In this there is only one group that is 300 randomly selected samples on basis of which her
survey will be done.
Second example:
· In this example, researcher decides to carry out a study that is the actual experiment is
carried out by giving treatment to the group.
· In this example, the selection of sample is random but with different method the
sample is selected that is researcher lists everyone’s names on chits, and draws one chit
out of the lot and writes that name under a column named linear CAI and puts that chit
aside.
· In this, there are two groups since there are two different CAI packages of which
effectiveness has to be checked
Friday, April 6, 2007
Simple random sampling v/s Convience sampling
Differences between simple random sampling and convenience sampling
Simple random sampling
1. In simple random sampling the final results can be generalized to the population.
2. In this every individual has an equal chance of getting selected from sampling frame
3. Selection of sample is without bias
Convenience sampling
1. In convenience sampling results cannot be generalized to the population.
2. In this type of sampling only those individuals are chosen who are easily accessible and does not have an equal chance of getting selected
3. Selection of sample may have biasness
Simple random sampling
1. In simple random sampling the final results can be generalized to the population.
2. In this every individual has an equal chance of getting selected from sampling frame
3. Selection of sample is without bias
Convenience sampling
1. In convenience sampling results cannot be generalized to the population.
2. In this type of sampling only those individuals are chosen who are easily accessible and does not have an equal chance of getting selected
3. Selection of sample may have biasness
Wednesday, April 4, 2007
Effective communication
The best way to engage in a real-world conversation is to go through the following five steps for effective communication: listen, understand value, interpret, and contribute.
Listen: Listening is like being a sponge, and the best sponges hold water indefinitely. Until you are ready to contribute—to squeeze some knowledge from your sponge—you need to be taking in a lot more that you’re putting out.
Understand: By understanding what is actually being said, apart from any biases or agendas—especially your own—you begin to value feedback. You need to ensure that you keep that value. Value the conversation, the individual, and the feedback more than you value your own opinion. If you don’t do this, when it comes time to contribute, your comments will be out of context and will hold much less value than they otherwise would.
Value: Valuing everyone’s contribution can be difficult in the best of times—some people in any large conversation don’t listen, don’t value others’ contributions, and therefore simply don’t deserve to be talking. However, when you’re a business listening to feedback about your company, products, and industry, it’s far too easy to discount certain contributions as unworthy of your attention. Don’t fall into this trap. Before you can contribute and properly respond to what’s going on in a conversation as big as the blog posting, you need to value everyone involved—after all, the one person you value one time could well be your next big customer evangelist.
Interpret: Before you take the step in becoming involved in the global conversation happening on blogs, you need to interpret and evaluate what has already been said and determine whether you actually have any valuable and unique insight to offer. After all, if the only thing you have to say in a large conversation is “Yes, I agree!”, it’s probably best to live by the adage, “Even a fool is thought wise if he keeps silent.”
Contribute: The final step in effective communication is to contribute something of value to the group. What valuable information can you offer? When the conversation centers on your area of expertise, you can offer authority, passion, and a unique perspective. Unlike most parties, where not everyone gets a chance to talk to everyone else, thousands of blog readers and writers are waiting eagerly to hear what you and your company have to say. Once you have properly prepared to contribute to the conversation, you can be sure that you will not only be heard, but that you will get feedback.
http://www.e-articles.info/e/a/title/The-Five-Steps-of-Effective-Communication-on-Blogs/
Listen: Listening is like being a sponge, and the best sponges hold water indefinitely. Until you are ready to contribute—to squeeze some knowledge from your sponge—you need to be taking in a lot more that you’re putting out.
Understand: By understanding what is actually being said, apart from any biases or agendas—especially your own—you begin to value feedback. You need to ensure that you keep that value. Value the conversation, the individual, and the feedback more than you value your own opinion. If you don’t do this, when it comes time to contribute, your comments will be out of context and will hold much less value than they otherwise would.
Value: Valuing everyone’s contribution can be difficult in the best of times—some people in any large conversation don’t listen, don’t value others’ contributions, and therefore simply don’t deserve to be talking. However, when you’re a business listening to feedback about your company, products, and industry, it’s far too easy to discount certain contributions as unworthy of your attention. Don’t fall into this trap. Before you can contribute and properly respond to what’s going on in a conversation as big as the blog posting, you need to value everyone involved—after all, the one person you value one time could well be your next big customer evangelist.
Interpret: Before you take the step in becoming involved in the global conversation happening on blogs, you need to interpret and evaluate what has already been said and determine whether you actually have any valuable and unique insight to offer. After all, if the only thing you have to say in a large conversation is “Yes, I agree!”, it’s probably best to live by the adage, “Even a fool is thought wise if he keeps silent.”
Contribute: The final step in effective communication is to contribute something of value to the group. What valuable information can you offer? When the conversation centers on your area of expertise, you can offer authority, passion, and a unique perspective. Unlike most parties, where not everyone gets a chance to talk to everyone else, thousands of blog readers and writers are waiting eagerly to hear what you and your company have to say. Once you have properly prepared to contribute to the conversation, you can be sure that you will not only be heard, but that you will get feedback.
http://www.e-articles.info/e/a/title/The-Five-Steps-of-Effective-Communication-on-Blogs/
Monday, April 2, 2007
Criteria for good sampling design
1st picture is of measurability
2nd picture is of economy
Criteria for a good sampling design:
1. Goal orientation
The sample design should be such that the survey research objectives must be met at all levels, it should be based on the studies goal and objectives.
2. Measurability
The sampling design provides the data necessary for analysis and is computed through statistical test
3.Practicality
The sample design must translate theoretical sampling models, in to clear, simple practical and complete instructions for the conduct of the survey.
4. Economy
The sample design should allow the survey objectives to be achieved at minimum cost, i.e. the objectives should be met with available time, financial, personal and any other necessary resources.
2nd picture is of economy
Criteria for a good sampling design:
1. Goal orientation
The sample design should be such that the survey research objectives must be met at all levels, it should be based on the studies goal and objectives.
2. Measurability
The sampling design provides the data necessary for analysis and is computed through statistical test
3.Practicality
The sample design must translate theoretical sampling models, in to clear, simple practical and complete instructions for the conduct of the survey.
4. Economy
The sample design should allow the survey objectives to be achieved at minimum cost, i.e. the objectives should be met with available time, financial, personal and any other necessary resources.
Subscribe to:
Posts (Atom)