Linkert scale first developed by Rensis Linkert in 1932 in measuring attitudes. This scale uses only items that are definitely good and definitely bad. tem is definitely favored, preferred, which is good, given the negative sign (-). The total score is the sum score of the response from the respondents that the results are interpreted as the position of the respondent. This scale uses ordinal measures so as to make the rankings despite not known how many times a respondent is better or worse than other respondents.
The procedure in making linkert scale is as follows:
The collection of items that are sufficiently numerous and relevant to the matter under investigation, in the form of preferred items that are bright enough and bright enough not favored. The items are tested to a group of respondents is fairly representative of the population who want to study. Collecting response from the respondent to then be given a score, to answer that gives an indication of favor given the highest score. The total score of each individual is the sum of the scores of each item of individual.
Response is analyzed to determine which items are very real boundaries between high scores and low scores in the total scale. To maintain internal consistency of the questions the items that showed no correlation with the total score or show no real difference whether the entry into the high or low scores discarded.
Excess linkert scale:
1. In drawing up the scale, items that obscure the correlation can still be included in the scale.
2. It's easier to make than the Thurstone scale.
3. Having a relatively high reliability than Thurstone scales for the same item number. Can also show items that are expressed in several alternative response.
4. Can provide more real information about income or attitudes of respondents.
Weaknesses linkert scale:
Only individuals can sort of scale, but can not compare how many times an individual is better than other individuals. Sometimes a total score of an individual does not provide a clear meaning, many patterns of response to some of the items will give the same score. The validity of the scale linkert still require empirical research.
The procedure in making linkert scale is as follows:
The collection of items that are sufficiently numerous and relevant to the matter under investigation, in the form of preferred items that are bright enough and bright enough not favored. The items are tested to a group of respondents is fairly representative of the population who want to study. Collecting response from the respondent to then be given a score, to answer that gives an indication of favor given the highest score. The total score of each individual is the sum of the scores of each item of individual.
Response is analyzed to determine which items are very real boundaries between high scores and low scores in the total scale. To maintain internal consistency of the questions the items that showed no correlation with the total score or show no real difference whether the entry into the high or low scores discarded.
Excess linkert scale:
1. In drawing up the scale, items that obscure the correlation can still be included in the scale.
2. It's easier to make than the Thurstone scale.
3. Having a relatively high reliability than Thurstone scales for the same item number. Can also show items that are expressed in several alternative response.
4. Can provide more real information about income or attitudes of respondents.
Weaknesses linkert scale:
Only individuals can sort of scale, but can not compare how many times an individual is better than other individuals. Sometimes a total score of an individual does not provide a clear meaning, many patterns of response to some of the items will give the same score. The validity of the scale linkert still require empirical research.
{ 0 comments... read them below or add one }
Post a Comment