LIKERT Likert five-point scale method

Likert scale method is a method to measure people's attitudes towards advertisements, products and other objects by using a compiled scale.

the compiling method of Likert scale was put forward by Rensis A．Likert in 1932. The process of compiling the scale can be divided into the following four steps:

Step 1: Draw up several statements about attitude objects. There are positive and negative attitudes expressed in these statements, and the answers of each statement are the same, all of which are five (or seven) grades. For example: strongly agree, agree, undecided, disagree, strongly disagree.

Step 2: Divide all statements into positive statements (such as "This brand is to my taste") and negative statements (such as "This brand is cold"). The answers to these two types of statements are given different scores. The scoring method of positive attitude sentences is: 5 points for strongly agreeing, 4 points for agreeing, 3 points for undecided, 2 points for disagreeing, and 1 point for strongly disagreeing; Negative sentences are given in the opposite way: agree with 1, agree with 2, disagree with 3, disagree with 3 and disagree with 4.

Step 3: Select a number of respondents and ask them to rate each sentence listed according to their attitudes and opinions.

Step 4: Choose discriminating sentences to form a formal scale. There are usually two ways to choose sentences: average difference method and internal consistency method.

the average difference method is to first change the answers made by respondents to each sentence into component numbers, and then arrange all respondents according to their total scores from high to low, with 25% of the intercepted top scores as high groups and 25% of the intercepted bottom scores as low groups. Find the average value of each sentence in these two groups, and take the difference between the average values of high and low groups as the standard of sentence screening. If the difference is large, it means that the sentence has strong discrimination ability, so it is selected; If the difference is small, it means that the sentence discrimination is poor, so it is excluded. The selected sentences can form the scale.

the internal consistency method is to arrange the total score of each promisor in one column and the score of a sentence in another column. If the number of sentences is large, directly calculate the rank correlation of the two columns of data; if the number of sentences is small, subtract the score of the sentence from the total score of the respondent, and then calculate the rank correlation. If the correlation coefficient is large, it means that the respondent's attitude towards the sentence is consistent with the overall attitude, so the sentence is selected. On the contrary, if the correlation coefficient is small, it means that the attitude of the sentence is inconsistent with the overall attitude, so the sentence is eliminated. According to this method, each sentence is filtered, and finally all the selected sentences can form a scale.

after the scale is made, its use method is: let the respondents reply to each sentence, then convert it into scores and add them up, so that you can get the attitude score of each respondent, and average the scores of all respondents to get the overall attitude of the respondents to the evaluated object. If these respondents are representative, we can infer the attitude of ordinary consumers.

In this scale, people's attitudes towards these issues are no longer simply agreed or disagreed, but divided into several categories, ranging from very much in favor to very much in disapproval, with a neutral category in the middle. Due to the increase of types, people's differences in attitudes can be fully reflected. In addition, because it is relatively simple, the testee can save time to complete it. Therefore, Likert scale is one of the most popular and commonly used scales in the study of opinions or attitudes.