The magnitude of the sample must be calculated based on the specific techniques that apply to the conclusion that the population can be accounted for. In addition it must also satisfy the sampling technique as described in Example retrieval techniques (sampling technique). The amount of sample members are chosen based on considerations: practicality, accuracy, nonrespon, and data analysis.
Practical Considerations Practical considerations related to elements of cost, time, energy and ability for exploration (exploratory) or the discovery and exploration, then the sample members do not need a lot but for eksplanari (explanatory) or explained, then the members of the sample had more;if we choose a lot of members of the sample, then the level is relatively accurate prediction, and calculate the error menstabulasi big, big realiabilitas, and power increased, vice versa.Accuracy The smaller we choose the level of significance or alpha (α), the more members of the sample.
Thus the more precise or thorough our expectations.NonresponConsideration Nonrespon consideration is the approximate number of samples which can be made after all members of the sample of respondents reduced the number of samples taken as a group test research instrument. Members of the samples that have been used as test groups should not be used as a respondent to obtain the actual data. In addition to the above considerations, also need to consider the amount respomden willing to return the questionnaire or be interviewed and observed.
Data Analysis Analysis of the data used to determine the amount of sample members. For parametric statistical techniques require a relatively large data (minimum 30), while for the data were analyzed with nonparametric statistical technique, simply use the data that is relatively small.
Practical Considerations Practical considerations related to elements of cost, time, energy and ability for exploration (exploratory) or the discovery and exploration, then the sample members do not need a lot but for eksplanari (explanatory) or explained, then the members of the sample had more;if we choose a lot of members of the sample, then the level is relatively accurate prediction, and calculate the error menstabulasi big, big realiabilitas, and power increased, vice versa.Accuracy The smaller we choose the level of significance or alpha (α), the more members of the sample.
Thus the more precise or thorough our expectations.NonresponConsideration Nonrespon consideration is the approximate number of samples which can be made after all members of the sample of respondents reduced the number of samples taken as a group test research instrument. Members of the samples that have been used as test groups should not be used as a respondent to obtain the actual data. In addition to the above considerations, also need to consider the amount respomden willing to return the questionnaire or be interviewed and observed.
Data Analysis Analysis of the data used to determine the amount of sample members. For parametric statistical techniques require a relatively large data (minimum 30), while for the data were analyzed with nonparametric statistical technique, simply use the data that is relatively small.
Tidak ada komentar:
Posting Komentar