We collected sensory scores from 30 semi-trained panels using hedonic scale (1-5) to evaluate a food product. So we need a suggestion to use more appropriate statistical software. Thank you.
You can use Microsoft Office Excel, SPSS, R etc. But, How many parameters do you have and, which you will do analyzes. You have to identify him. It's more important.
30/5000
Firstly, you need to do a homogeneity test to check your data. You should see if there is a significant difference between the variances of these parameters. You should do this, for example, if you do it weekly. If there is a difference, you can do multiple comparisons to find the cause. For example, you can use ANOVA to test the difference. Then you can make DUNCAN for multiple comparisons for difference.
I think excel start is the best because the software will provide detail information about the analysis.
I have done sensory analysis using IBM SPSS statistic software for 5 parameters and for me it was kind a easy to use.
Have used Statgraphics in the past using principle components analysis. Now using Minitab
Dear brother,
According to me in sensory evaluation the number .of samples you want to analyze are not important, but the number of groups should be considered closely. if the number of groups are two you can use independent sample T test (in case of parametric parameters are achieved) or its alternative nonparametric test. If the number of groups exceed from two groups the one-way ANOVA or its nonparametric alternative Kruskal Wallis test with multiple comparision tests (MCT) for example Tukey, Duncan, etc... should be used.
The SPSS software seems better for its convenience and accuracy.
With best regards
If you used the hedonic scale to rate 7 attributes of a food product you need to report the mean and standard deviation of each attribute. In this case, Excel will do the job very well. On a related matter, the 9-point scale is the most-used in hedonic testing as it provides effective discrimination when working with a number of samples. Cheers,
ANOVA to test the statistical difference i.e. significantly differ of each attribute among the headonic test(0-9) . Then you can make DUNCAN for multiple comparisons for difference.
The first thing you need to ask yourself is what kind of information do I need from the sensory data? The answer to this question will guide you in the choice of software to use. If you want to compare between the mean of the data, u can use XLSTAT or SPSS. But if it goes beyond comparison in which u want to know the relationship between each attribute or sample or else analysis the panelist or conduct multivariate analysis on the data, I will suggest you use XLSTAT or The Unscrambler (Majorly for multivariate analysis ). Finally if u want to conduct a multi-way (three way) analysis in which u analyse both the panelist, samples and attribute at the same time using parallel factor analysis, N-PLSR or Tucker 3, I will suggest you use Matlab, R, or Solo 862 (by eigenvector). Thanks I hope it helps
Bilatu Agza Gebre , In respect to your question as to which software package would be efficient for sensory data analysis, i recommend that you use XLSTAT.
Although your sample size matters, the analyses you wish to apply to your data would depend on the replicates to a less extent. For a sample size of 30, with at least 2 replicates, if you wish to decipher the relationship between the attributes or samples based on their means and SD, I would suggest SPSS, however, conducting multivariate analysis, i would strongly suggest XLSTAT over Minitab simply for ease of use.
Usualli i use SPSS, Statistica, Excel, Origin. With programming functions and many libraries - R software/
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software