Arabic:7.5818
French :3.5636
English:4.4723
I'm working on attitudes towards Arabic, French and English as a meduim of instruction among students at University. and I got these results. My sample number = 55
I understand that Arabic is the highest rank then comes English and French at last. But my question is how can I interpret that data inside my text.. I read some articles where researchers have specified the mean scores of the results. For instance, one guy mentioned that low scores is between range 1.0-2.4 meduim between range 2.5-3.4 and high is between range 3.5-5.0. ( the researcher followed Oxford & Burry-stock (1995) scales
Hello,
Before any interpretation is possible you need to know what is the range of scores? Is it uniform across all three languages? With this information mean scores are more meaningful than on their own. Also it is advised to report standard deviations.
Regards, Witold
I would recommend using box plot (a.k.a. box and whisker diagram) in your analysis.
Davide Giacalone made a good point: you need to measure uncertainty, and box plots will help you analyse the distribution of scores.
As for the interpretation of the level of scores - I think you just need to follow Oxford & Burry-stock (1995) scales if you used the same technique to evaluate the scores.
Thank you all for your valuable guide. I tried to solve the mess I made, Can you still check what I did ? Can I now say that Arabic is the Highest and French is the lowest?
The Likert scale of three points (A/NAND'DA) consisted of 16 questions, the questions that deal with this main point are 2 for each language . Cronbach's Alpha for each of the three languages that investigate students attitudes towards them as a tool of instruction was in all higher then .5 . 55 respondents answered the questionnaire.
Thank you in advance.
I think now the task is to formulate your hypotheses.
The means seems to be different, but the distributions of scores are not Gaussian. So including box plots would give more information.
Huum, never used it before and dunno how to use it?
Isn't there any other easier way ?
Or can I just go ahead and explain which is high and which is low by score of means?
I appreciate your efforts and time Andrey.
Thank you.
You can, of course, just use the means, but someone reading your results can be curious of how the scores are distributed among students. It is important to know the empirical distribution of your data because there can be several outliers that influence the sample mean. These unusual observations can influence the conclusions of your analysis. E.g., if most of students speak Arabic fluently, but there's a group (say, 15%) of students that do not speak Arabic at all, how can you use it as a medium of instruction?
Box plot is s relatively simple thing: https://en.wikipedia.org/wiki/Box_plot
Great. I got your point. They all speak or have at least a basic knowledge in the languages used: Arabic (native) French( 2nd lge) and English (3rd foreign lge).
Thnx a lot :) and thnx for the link
I appreciate it .
Dear Soumia,
There's one factor that you have to mention to those who are trying to help you interpret your data: it's the fact that the Arabic you're talking about is NOT the students' mother tongue (diglossia relationship). That is, MSA as the learners' school language might be, in certain cases at least, as 'distant' from their Arabic vernaculars as Standard French is from Haitian or from other French-based creoles, or as Latin is from Modern Italian !
Dear Dr.
Yes,That is right. I basically wanted help with SPSS operations and was running out of time.
SPSS can help you when you study is well. Secondly, study how to calculate mean range. It is very important. Some people the mean range of 2.4 below and 2.5 high. I have a result now with my mean range 1.0185 as high and the percentage of the data is 98.2%. This is another mean range of what other people are using.
The higher the mean score the higher the expectation and vice versa. This depends on what is studied.E.g. If mean score for male students in a Mathematics test is less than the females, it can be interpreted that female students perform better than the male students in the test.
your interpretation should depend on the data type you gathered, however, there is a debate that likert scale data could not be presented using mean scores, rather median is recommended; therefore, you need to clear that which type of data scale you had before going to interpreting the, mean.
assuming a normal distribution, can you give us the standard deviation of each group?
also , as was said before: what are your hypotheses and what is your study design?
help me interpret the following budgetary characteristics
FINANCIAL/YEAR
FY 2013/14
FY 2014/15
FY 2015/16
FY 2016/17
FY 2017/18
mean
1.376
1.25
1.5
1.358
1.252
min
1.169
2.663
1.02
0.768
0.532
max
2.063
1.61
1.72
5.66
3.55
std dev
0.181
0.112
0.127
0.789
0.58
skewness
1.743
0.546
1.19
4.518
2.866
kurtosis
1.169
2.663
1.02
0.768
0.532
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS