Hi, I am trying to compare whether there is a difference in students' attitude from different types of school. The survey consists of their school type and 21 questions on attitude in 4-point likert scale using google form. 100 respondents have completed it and I have already input all the data to spss. Since there are many ways to analyse likert scales, I'm wondering which method is most suitable for my research? I would also like to know 1) how do I compute the means of several questions into a subscale, if possible? 2)I have tried t-test, however it says a. t cannot be computed because at least one of the groups is empty. Could someone please explain in detail for me because the deadline of submission is approaching . Thank you.

0 |

7/18/2016 6:48:32 AM

On Sun, 17 Jul 2016 23:48:32 -0700 (PDT), Hannah <hanakee311@gmail.com> wrote: >Hi, >I am trying to compare whether there is a difference in students' attitude from different types of school. >The survey consists of their school type and 21 questions on attitude in 4-point likert scale using google form. 100 respondents have completed it and I have already input all the data to spss. > >Since there are many ways to analyse likert scales, I'm wondering which method is most suitable for my research? A Likert Scale is the sum (or the average, to preserve the meaningful anchor points) of a number of Likert items. Compare means with ANOVA (which, for two groups, is a t-test). > >I would also like to know >1) how do I compute the means of several questions into a subscale, if possible? COMPUTE ATTscore= mean.10(v1 to v21). >2)I have tried t-test, however it says a. t cannot be computed because at least one of the groups is empty. "At least one of the groups is empty)". Believe it. If you do not specify the scores for the two groups, T-TEST assumes that they are coded 1 and 2. > >Could someone please explain in detail for me because the deadline of submission is approaching . Google for the SPSS tutorials available from UCLA. -- Rich Ulrich

0 |

7/18/2016 8:27:44 PM

On Monday, July 18, 2016 at 4:27:48 PM UTC-4, Rich Ulrich wrote: > On Sun, 17 Jul 2016 23:48:32 -0700 (PDT), Hannah > <hanakee311@gmail.com> wrote: >=20 > >Hi, > >I am trying to compare whether there is a difference in students' attitu= de from different types of school. > >The survey consists of their school type and 21 questions on attitude in= 4-point likert scale using google form. 100 respondents have completed it = and I have already input all the data to spss. > > > >Since there are many ways to analyse likert scales, I'm wondering which = method is most suitable for my research? >=20 > A Likert Scale is the sum (or the average, to preserve the meaningful=20 > anchor points) of a number of Likert items. Compare means with ANOVA > (which, for two groups, is a t-test).=20 Here's a note on several distinctions that people are often quite careless = about making. The OP (Hannah) may find it useful. >=20 > > > >I would also like to know > >1) how do I compute the means of several questions into a subscale, if p= ossible? >=20 > COMPUTE ATTscore=3D mean.10(v1 to v21).=20 That .10 in Rich's example instructs SPSS to compute the mean of variables = v1 to v21 only if there are at least 10 variables with valid data. If that= condition is not met, it returns a system missing value. If there is some= other minimum number required, change .10 to something else. --- snip the rest --- HTH.

0 |

7/18/2016 9:45:49 PM