Lab 16 Instrumental versus Sensory Analysis of Texture Flashcards
Describe the double compression test.
- Designed to mimic the force needed to chew and breakdown food
- Data can be used to determine a wide range of physical parameters
Describe the triangle test.
- 2 samples are the same and 1 is different
- Panelists are asked which sample tastes different from the other two
- Panelists may also be asked to describe what the difference was (e.g., sweeter, saltier, etc.)
What is a continuous dependent variable?
- An infinite set of values within an interval (e.g., pH - 2.66, 4.59, 6.71, etc.)
What are nominal values?
- Values that have no order (or ordering is not equally spaced)
- e.g., yes/no, correct/incorrect
What are ordinal values?
- Values can be ranked/have natural order
- e.g., Likert scale: strongly dislike, neutral, strongly like
What is a parametric test?
- Assume underlying statistical distribution (e.g., normal distribution, equal variance, equal sample sizes)
- Used when dependent variable is continuous
- e.g., t-test, ANOVA, Pearson correlation
What is a non-parametric test?
- Do not assume that data are normally distributed
- Use ranked data rather than actual data values (since actual data is not used tests have less statisical power)
- Must be used when dependent variable is ordinal or nominal (can also be used for continuous data if assumptions are violated)
Why do non-parametric tests have less statistical power than parametric tests?
- Since actual data is not used (ranked data is used)
For the following non-parametric test, describe the independent variable, the dependent variable, and an equivalent parametric test.
Wilcoxon signed-rank test
- Independent variable: categorical with two related groups
- Dependent variable: ordinal or continuous but not normally distributed
- Equivalent parametric test: paired t-test
For the following non-parametric test, describe the independent variable, the dependent variable, and an equivalent parametric test.
Mann-Whitney
- Independent variable: categorical with two unrelated groups
- Dependent variable: ordinal or continuous but not normally distributed
- Equivalent parametric test: independent t-test
For the following non-parametric test, describe the independent variable, the dependent variable, and an equivalent parametric test.
Kruskal-Wallis
- Independent variable: categorical with 3 or more unrelated groups
- Dependent variable: ordinal or continuous but not normally distributed
- Equivalent parametric test: one-way ANOVA
For the following non-parametric test, describe the independent variable, the dependent variable, and an equivalent parametric test.
Friedman test
- Independent variable: categorical with 3 or more related groups or with 2 independent variables
- Dependent variable: ordinal or continuous but not normally distributed
- Equivalent parametric test: repeated measures ANOVA and two-way ANOVA
What does a monotonically increasing function look like?
Name 3 non parametric correlation tests.
- Spearman correlation
- Chi-square test of association
- Binomial distribution
Describe Spearman correlation.
- Tests the strength of a monotonic association
- Variables can be continuous or ordinal
- Performs Pearson correlation on ranked data
Describe the chi-squared test of assocation.
- Tests the strength of a relationship between two categorical variables
- Variables must have two or more groups
- For a triangle test the two variables are incorrect/correct and observed/expected
How can objective data regarding the physical properties of food be measured?
- Texture profile analysis: double compression test
- Samples are compressed twice using a texture analyzer to provide insight into how samples behave when chewed.
- Often called ‘two bite test’.
Given this TPA output, how can you calculate springiness?
Length 2/length 1
Given this TPA output, how can you calculate cohesiveness?
Area 2/ Area 1
Given this TPA output, how can you calculate chewiness?
Hardness x Springiness x Cohesiveness
