Meauring Sensorial

You might also like

You are on page 1of 16

MEASURING SENSORIAL QUALITY OF IBERIAN HAM BY

RASCH MODEL
C. GARCIA J. VENTANAS’, T. ANTEQUERA’, J. RUIZ’, R. CAVA’
and P. ALVAREZ’

’Tecnologia de 10s Alimentos, Facultad de Veterinaria, Universidad de Extremadura


’Economia Aplicada y Organizaci6n de Empresas, Facultad de Economicas y
Empresariales, Universidad de Extremadura

Accepted for Publication September 9, 1995

ABSTRACT

This paper focuses on afresh way to determine quality of ham according to


sensorial analysis. It is an application of latent traits theory to biometry. “Qualiw
of Iberian Ham” can be considered a latent variable defined by a set of sensorial
analysis factors (items). The theoretical background is Item Response Theory
(IRT), which suggests that ifwe can understand how each item in a set of items
operates with an object, then we can estimate a measurefor the object. The Rasch
model is the most common modelfor that theoty. This technique has been applied
to datafiom 8 dgerent hams assessed by 15 expertjudges tasting ham in order to
obtain Rasch measurementsfor hams and calibrationsfor the items.

INTRODUCTION

Latent trait models, in test theory, focus on the interaction between a person and
an item rather than upon test scores. The mathematical formula models a response
to an item. The most representative model for Item Response Theory is the Rasch
model. It is an instrument for measuring latent traits (Andrich 1988). This method
has been used in other fields as the study of the monofloral honeys of Extremadura
(Lozano 1993) the sensorial characteristics of wine (Alvhez 1992) and in
comparison with Principal Component Analysis (Horimoto et al. 1995).

3T0whom correspondence should be addressed: Tecnologia de 10s Alimentos, Facultad de


Veterinaria, Avda. de la Universidad s/n 10071 Caceres, SPAIN; Phone 34 27 25 71 22 Fax
number 34 27 25 71 10

Journal of Food Quality 19 (1996). 397-412. All Rights Reserved.


OCopyright 1996 by Food & Nutrition Press, Inc., Trumbull, CT 0661I 397
398 C. GARCIA, J . VENTANAS, T. ANTEQUERA, J. RUIZ, R. CAVA and P. ALVAREZ

Measurements are based upon observations which are essentially qualitative.


To make the measurements we develop rules which control how these observations
are formed. These rules specify the degree of validity and accuracy that we want
for the measurements.
The Iberian ham is an uncooked, salted, dried and high quality meat product,
produced in the southwest of Spain from Iberian pigs (Ventanas er al. 1991). By
regarding the feeding, which the pig has been given before the slaughter, the
following differences are made: (1) Acornfed pigs or pigs whose feedind cycle
terminates on mountain pastures. (2) Fodder-fed pigs or pigs whose feeding cycles
terminates on fodders. The Iberian hams produced from acomfed pigs are referred
in the market as higher quality than hams from fodder-fed pigs. Sensory profiling
of these types of ham have been used to establish a detailed qualitative and
quantitative specificationof their sensory characteristics. The aim of this paper is
to find out a way for measuring quality in order to show the hams produced from
acornfed pigs are the best.
Measuring “quality of Iberian ham” can be analogous to measuring height.
First we bring to the fore our idea of the “quality of Iberian ham” variable that we
want to measure. Next, we determine which observationswill be useful to consider
as informativemanifestations of that variable. Then we choose the quality of ham
agents, which can manifest instances of this unidimensional “quality” variable.
The idea of a line upon which say “arithmetic” items are positioned provides us
with a picture of the “arithmetic” variable. This shows us how to proceed in the
construction of that variable. We use our knowledge of “arithmetic” to position
items along the line and their continuum to revise these positions by empirical
analysis (Guedes et al. 1995).

METHOD

The Rasch Model


“Quality of Iberian ham” can be considered as a latent variable (Xni)defined by
a set of items. These items ( 8 , ) are those criteria on which it is presumed that
“quality” depends. They allow ham to be assessed (p,) for “quality”. Like any
other variable, “quality of Iberian ham” is conceived as a line with direction, along
which items and hams are located (Fig. 1). It is assumed to be a simple dimension.
The direction implies “more” of the variable. More is “more” distance along the
line. This concept is needed to find ways of establishing the locations of items,
location being a helpful way of thinking about the quality of hams.
According to the assumption, the following picture segments the manner in
which ham assessment no. 1 (PI) and items 8 , can be thought of as located on the
quality line (Fig. la). Items 81,8*, 8 3 , in this case, are closer to the low end than
ham p, and item 8 , (Wright and Stone 1978). If there are two or more hams (fin),
MEASURING QUALITY OF HAM BY RASCH MODEL 399

differences in quality will be revealed by a set of items B I on the relative locations


related to items (Fig. lc). Then the latent trait “quality” is conceived as a continuum
along which parameters bi for agents and B, for objects are located.

a)
8,
I I I I

-.-
low quality 6, 62 63 64 high quality

b)
PI 6, 62 63 64
X X- X X- a ham B, would not expected to account

.-
for any quality criteria (item)

4 82 83 84 84
-X- X X
- X ham 4, would expected to account for
all quality criteria.

FIG. 1. (A) IN RASCH MODEL, THE QUALITY OF IBERIAN HAM IS CONCEIVED AS A


LINE IN WHICH HAMS ASSESSMENT (b) AND ITEMS (A,),CAN BE LOCATED. (B)
LOCALIZATION OF H A M S ASSESMENT IN FRONT OF OR BEHIND THE ITEMS
INDICATES THAT THE HAM WOULD NOT BE EXPECTED TO ACCOUNT FOR QUALITY
CRITERIA. (C) IF THERE ARE TWO OR MORE HAMS, DIFFERENCES IN QUALITY WILL
BE REVEALED BY ITEMS AND HAMS ASSESMENT LOCALIZATION
This picture shows the ham pI to have the lowest quality. Ham p2 surpasses only the quality item
tSl. Ham p1 surpasses the quality factor items tSl and The ham p3surpasses all these quality
items. p,, is the ham with the lowest quality and p., is the ham with the highest quality.

Let Xnibe the quality dichotomous variable which means that the ham “n” will
score on item ‘‘i’’. If the score is 1, that is X,,=l, then it is said that ham “n” has
some quality; other wise Xni=0.
Therefore: if B, -bi > 0 then P[X,,= 11 B0.5
if B, - b 1 < 0 then P[Xni= 11 <0.5
if 13, -bi = 0 then P[X,,= 1]= 0.5

This analysis allows us to relate the probability of having quality to the difference
(0, -4). This difference can range from --to +-; and probability range from 0 to
1, that is 0 I P{X,, = 1) i 1

--s (pn-bi) s+-


400 C. GARCIA, J. VENTANAS, T. ANTEQUERA, J . RUIZ, R. CAVA and P. ALVAREZ

If we use the difference as an exponent of base g, then

0se ’ m - b l < +m

With a further adjustment the following expression can be obtained and its
limit is

We take this formula to be an estimate of the probability of having quality when


a ham “n” score 1 on item “i” given the parameter p, and hi.The relation can be
written as

This is the formula that George Rasch (Rasch 1980) chose in his development of
latent trait theory.
The probability when Xni=0 is

Suppose R, = 6 and hi = 4 then

It is clear from the formula that it is unimportant that the value for B, was 6 and for
hiwas 4. The important thing is that they were two units apart. Any pair of values
differing by 2 will produce the same probability. Our scale is interval not ratio,
where the numbers chosen on the scale are arbitrary so long as the difference is 2.
Suppose ham “n” score on a set L of items X,,, Xn2,Xn3,....,XnLwhere each of the
elements will be 1 or 0. One way for accounting for “quality” will be the number
of items that ham “n” scores on items, ignoring the score pattern. We are aware
that not all items are scored in the same way for each ham; then the pattern of the
score is important. The total score which a ham obtains can be represented by r,
= X,, + Xn2+ .... + XnL,The conditional probability that if a certain total score is
MEASURING QUALITY OF HAM BY RASCH MODEL 40 1

obtained it will be obtained through a particular pattern of 1 and 0:

- “*fin]IPn,’J
P(rnI Pn,’-)
*
= ; where - (“Till”) indicates all values are referred to.

Thus, Xniindicates the score of ham “n” to all items referred to as “i“, goes from
1 to L. Thus, Xnirefers to the pattern of scores from ham “n”.
The numerator is the probability of scoring with a particular pattern and
obtaining the total score which that pattern yields. The denominator is the
probability of obtaining that total score by any pattern. It can be shown that

i
*=

The sum is taken over all values of “i”, that is, over “all items, given that the total
score for the pattern of scores involved is rn”. That restriction on the summation
is represented by v, I r,.
Then the probability of obtaining a particular score by one pattern rather than
another depends on the paramaters of the item; the pattern of scores provides no
information about the hams. The information is provided by the total score. Rasch
model is the only latent trait model which justifies the use of total score.
Suppose that 8 , = 1 and h2= 4, then the probability of the ham “n” scoring 6,,
and not B2 is

and the probability for scoring h2 and not b , , is

In a practical situation we do not know the locations of the items’ parameters ,but
we can observe the pattern of the scores of the hams on the items.
For example, suppose we have 100 hams assessed by two items in a
402 C. GARCIA, J. VENTANAS, T. ANTEQUERA, J . RUIZ, R. CAVA and P. ALVAREZ

dichotomous way. If 18 hams score on both items and 9 hams did not score, then
there is a total of 27 hams that provide no information, the situation tells us nothing
about the items. If from the remaining 73 which score on only one of two items we
found that 69 (95%) were scoring on one item (item i = I) and 4 (5 %) were scoring
on the other (item i =2) as Eq. 6 and 7 show, we would scale the items 3 logit apart.
We could say that 8 , , was 1 and was 4, or 8 , = 0 and 82 = 3, or b l = - 1.5 and ti2
= 1.5. Our scale is interval not ratio, so where the numbers are chosen on the scale
is arbitrary so long as the difference is 3.
We used r, as the total score (number of items scored by the ham “n”)
defined by

If we consider scores to an item, rather than the score to a ham, we could count the
number of hams scoring that item over all N hams. Then

N
S,=Cxni
i-1
(9)

We could show that, if a certain number of hams yield a score of an item, which
ones they are will not depend on the item. So, not only does the Rasch model
justify the use of a total score for hams but also for items. This does not mean that
the number m and Sj should be used as a measure, but it does mean that they
contain all the information which is needed to estimate parameters r, and Si. They
are sufficient statistics for estimation.

Maximum Likehood Estimation


Let us consider the full data matrix (NxL) of total scores for hams (N) scoring
all the items (L). The probability of the pattern observed scores having occurred
will be the product of the separate probabilities for all the scores of each ham to
each item.
This probability, the likehood (A) of occurrence of the data matrix, is given by
MEASURING QUALITY OF HAM BY RASCH MODEL 403

Taking natural logarithm and with further calculus it is obtained:

N L N L

n i n i

This equation expresses the log likehood of the observed pattern scores in terms of
the parameters p, ,li,Si and r,. The patterns of the individual scores of ham on
items do not appear, only the total. This, together with the separation of r, p, and
Si ii,establishes the sufficiency of r, for estimating p,, and of Si for estimating ii
and provides the mean of obtaining estimates of the p, which are independent of
the 4 and vice versa. This equation would allow us to calculate the probability of
occurrence of a complete matrix of scores if we know the parameters p, and &.
The best estimates of parameters p, and 4 are found by maximizing the
likehood function given by the Eq. 1 I . An initial set of estimates is taken and the
log likehood of occurrence is calculated using Eq. 1 I. The estimates are then
altered in a direction which will increase the likehood occurrence of the observed
data. This process is continued until the estimates of parameters (p, and hi) which
best account for the actual score pattern we have obtained.
Using calculus it can be shown that the likehood function is maximized when
for each ham “n”

i-1

and for each item

N
si=cP(X,=l)
n-1

Equation 12 actually represents N equations, since there is one for each ham.
Equation 13 actually represents L equations, since there is one for each of the L
items. There are N hams involved but they cannot all obtain different total scores,
unless there are more items than hams, if any ham does not score any items and
thus score r, = 0, p, cannot be estimated since they could be anywhere along the
quality line below the items. Similarly the hams which score all items r, = L cannot
be estimated. Furthermore all hams obtaining the same score in the range r, = 1 to
404 C. GARCIA, J. VENTANAS, T. ANTEQUERA, J. RUIZ, R. CAVA and P. ALVAREZ

(L-1)will be estimate to have the same parameter pr. We don't need to write p,
since it applies to all hams with the score r. Instead of one equation for every ham
only one equation for each of the (L-1) acceptable scores on L items is needed, that
is for scores from I to (L-I)the rest of Eq. 12 becomes

L
r = C P(X,=l)
i-I

Where P (Xri= 1) is the probability of hams scoring r on item i.


Equation I3 becomes:

L-1
si=cn p ( x r i = l )
r-1

Where n, is the number of hams obtaining a score of r.


With the (L-1)estimates pr and L estimates of 8i, the L(L-1) estimated
probabilities of a ham with a score of r on item i will be given by:

In the computer programs which derive estimates for the parameters starting values
for the (L-1)estimates associated with each acceptable score are taken as

p, =log)-- r 1
L-r

and the starting values for the L item estimates are taken as
MEASURING QUALITY OF HAM B Y RASCH MODEL 405

The term subtracted serves only to fix the mean of the starting values for biat zero.
The scale is only interval so the origin is arbitrary. Fixing the mean at zero simply
fixes the scale on which the relative positions of both the item and the ham are
located. Successively better estimates for the pr and b iare obtained until successive
estimates on the right hand sides of Eq. 14 and Eq. 15 move closer to the observed
totals by less than a very small amount. These are then the best estimates of the pr
and tii in the sense that with no other values would the actual scores obtained be
more likely to have occurred. The procedure described is a method for
unconditional estimation (Wright and Masters 1982).
The standard errors of the estimates of the items parameters are given by

The standard errors of the estimates of the hams parameters are given by

1
W P )=
L

In this case, quality is considered as a latent variable defined by the following


items which represent the quality criteria for a Iberian ham:

Yellow color of fat Tenderness of lean


Roseate color of fat Salty taste
Fluid aspect of fat Sweet taste
Red color of lean Bitter taste
Bright of lean Flavor intensity
Ham slice marbling After-taste
Odour intensity Meat cured flavour
“Acorn ham” odor Rancid flavor
Tenderness of fat Roast flavor
Fluidity of fat Nutty flavor
Dryness of lean Caramel flavor
Fibrousness of lean Mouldy flavor
Juiciness of lean “Cellar ham” flavor
406 C . GARCIA, J . VENTANAS, T. ANTEQUERA, J. RUIZ, R. CAVA and P. ALVAREZ

These 26 items will correspond to the bi (i= 1,2,3,..26)parameter for quality. The
156 assessment from 8 hams (5 fed from acorn and 3 from different kind of feeds)
tasted by 15 ham judges will correspond to the Bn (n= 1,2,3,..,156) parameter.
Scores of hams on items are a computation of the level of the '5'' quality criterion
of the "n" ham assessment.
The amount computed for each quality criterion level for each ham, assessed
by all judges, is expressed on a 1 to 10 scale (Alvarez et al. 1993).
Parameters pn and bi are estimated by the maximum likehood method for 10
categories using the PROX (Wright and Douglas 1977) and UCON (Wright and
Mead 1976) algorithms.

RESULTS AND DISCUSSION

We have constructed a quality variable and located items and hams along it
from our observations. The map of the variable is a picture of the extent to which
we have accomplished the task of variable construction.
Hams separation indicates how efficiently a set of items is able to separate those
hams measured. Item separation indicates how well a sample of hams is able to
separate those items used.
It is desirable to locate hams assessment and items (quality criteria) along the
variable line with sufficient precision to be able to see between them. The more
items and hams are separated along this line the more usable are their
measurements.
The distance among items also identifies the direction and meaning of the
variable (Table 1). The item locations are the operational definition of the variable
of interest while the hams assessment locations are the application of the variable
to measurement (Table 2).
The least raw score items are those that imply higher measurements (Fig. Ic),
they turn out to be items 25; 24; 23; 22; 17. Items with higher raw scores imply
lower measurements and they are items 9; 12; 1 1 . The item 9 calibration is smaller
than the item 25 (Table I). The same can be said for the hams assessment. The
highest quality ham is the assessment no. 41, which is identified by "judge8Ham-
Acorn4" which means the ham no.4 from acornfeed pigs it is tasted by judge no.
8. The lowest quality ham is the one on the assessment 147 (Table 2).

Item and Place Fit


When a Rasch analysis is made, the next logical step is to analyze the extent
to which each data of the ham on each item fits the Rasch model expectation.
MEASURING QUALITY OF HAM BY RASCH MODEL 407

TABLE I .
COMPUTATIONS OF MEASURE FOR ITEMS

YTRY RAY INFIT OUTFIT


NUN SCORE COUNT MEASURE ERROR NWSP STD NNSP STO PTBIS ITEMS

25 9a 156 56.5 .4 1.29 1.8 1.2~ 1.5 .33 Mouldy flavour


24 115 156 56.0 .4 1.46 2.9 1.14 .a .40 Caramel flavour
23 140 156 55.4 .3 1.21 1.5 1.07 .5 .3a Nutty flavour
22 211 156 54.0 .3 1.40 3.3 1.48 3.7 .25 Roast flavour
17 218 156 53.9 .3 1.47 3.9 1.57 4.4 .21 B i t t e r taste
26 295 156 52.7 .2 1.74 6.7 1.56 4.9 .55 "Cellar ham" flavour
1 368 156 51 .a .2 1.09 .9 1.09 .a .41 Yelow colour of f a t
21 426 156 51.2 .2 1.21 2.1 1.24 2.3 .3a Rancid flavour
16 482 156 50.6 .2 .94 -.6 .94 - .5 .34 SYeet taste
2 516 156 50.2 .2 1.12 1.1 1.14 1.4 .33 Roseate colour of f a t
5 610 156 49.3 .2 .60 -3.9 .59 -3.9 .6a Bright of lean
a 615 156 49.2 .2 .76 -2.3 .77 -2.2 .71 "Acorn ham" d o u r
3 626 156 49.1 .2 .TI -2.2 .76 -2.2 .67 F l u i d aspect of f a t
19 656 156 48.8 .2 .57 -4.0 .57 - 4 . 0 .69 After-taste
7 671 156 48.6 .2 .66 -3.2 .67 -3.1 .66 Odour intensity
18 679 156 48.5 .2 .61 -3.7 .60 -3.7 .66 Flavour intensity
20 680 156 48.5 .2 .a3 -1.6 .a5 -1.3 .61 Meat cured flavour
6 687 156 48.5 .2 1.02 .2 1.01 .1 .45 Ham s l i c e marbling
4 704 156 48.3 .2 .94 -.5 .96 - .3 .34 R e d colour of lean
10 727 156 48.0 .2 .94 -.6 .93 ~ .6 .55 F l u i d i t v of f a t
t.9 ',.08 Tenderness of Lean
14 756 156 47.7 .2 1.18 1.6 1.21
15 769 156 47.5 .2 1.11 1.0 1.16 1.4 .20 Salty taste
13 779 156 47.4 .2 .60 -3.6 .60 -3.6 .60 Juiciness of lean
11 a3a 156 46.7 .2 1.26 2.3 1.26 2.3 .09 Dryness of lean
12 849 156 46.6 .2 1.33 2.9 1.32 2.8 .03 Fibrousness of lean
9 967 156 45.0 .3 1.25 2.1 1.30 2.6 .37 Tenderness o f f a t

MEAN
S.D.
557. 156.
243. 0.
50.0
3.0
.2 1.05
.Ol .31
.3 1.04
2.71 .30 .?:I I
MFIT is a standardized information-weightedmean square statistic, which is more sensitive to
unexpected responses to items near the ham location quality level.
MNSQ is the mean-square infit statistic, with expectation 1. Values substantially less than 1
indicate dependency in the data; values substantially greater than 1 indicate noise.

OUTFIT is a standardized outlier-sensitive mean square fit statistic, more sensitive to unexpected
ham score on items far from the hams location level.
MNSQ is the mean-square outfit statistic, with expectation 1. Values substantially less than 1
indicate dependency in the data; values substantially greater than 1 indicate the presence of
unexpected outliers.
PTBIS is the point-biserial correlation between the individual item score and the test person score
for the scored observations used in the analysis. Negative values for items often indicate missing
scores.

An item fit statistic is calculated for each item. This summarizes the extent to
which the pattern of the data of the sample on that item is consistent with the way
these hams have data on the other items. This gives a consistency fit statistic for
each item and for each ham assessment, and also for any subsets of items and hams
which might interest us.
408 C. GARCIA, J. VENTANAS, T. ANTEQUERA, J. RUIZ, R. CAVA and P. ALVAREZ

TABLE 2.
COMPUTATION OF MEASURE FOR HAMS AND JUDGES

YTRY RAU INFIT OUTFIT


NUM SCORE COUNT MEASURE ERROR MUSO STD MNSO STO P T B I S JUOGE,NUM/HAM,TYPE,NUU
41 1 42 26 51.3 .b .67 -1.2 .64 -1.3 .79 judge8HamAcornl
51 142 26 51.3 .6 2.11 4.0 2.01 3.6 .75 judge8HamAcorn5
9 139 26 51.1 .6 1.10 .4 1.07 .3 .MI judgel0HamAcornl
17 13P 26 51.1 .6 1.16 .6 1.18 .7 .67 judge8HamAcorn2
129 137 26 50.9 .6 .86 -.5 .82 -.6 .74 judge8HamAcorn4
55 135 26 50.7 .b 2.15 4.2 2.31 4.7 .59 judgel3HamAcorn5
6 133 26 50.6 .6 1.42 1.5 1.48 1.7 .47 judge7Ham3Acornl
52 132 26 50.5 .6 1.35 1.3 1.33 1.2 .60 judge9HamAcorn5
7 131 26 50.4 .6 1.43 1.6 1.42 1.5 .52 judge8HamAcornl
20 131 26 50.4 .6 1.40 1.7 1.56 2.0 .63 judge1 lHamAcorn2
22 131 26 50.4 .6 1.57 2.1 1.57 2.1 .35 judgel3HamAcorn2
21 131 26 50.4 .6 1.74 2.7 1.78 2.8 .59 judge5Acorn3
31 131 26 50.4 .6 1.02 .1 1.01 .1 .72 judgc9Acorn3
120 128 26 50.2 .6 1.10 .4 1.00 .o .77 judgeBHamAcorn3
19 126 26 50.0 .6 1.62 2.2 1.86 3.1 .29 judgelOHamAcorn2
30 126 26 50.0 .6 1.14 .5 1.06 .2 .67 judge8HamAcori-G
32 126 26 50.0 .6 .88 -.4 .84 -.6 .81 judgelOHamAcorn3
109 124 26 49.9 .6 1.19 .7 1.17 .6 .67 jud~elOHamAcorn2
141 124 26 49.9 .6 1.20 .7 1.16 .6 .67 judge8HamAcorn5
131 122 26 49.7 .6 1.89 .2 1.99 3.5 .37 judgelOHem4corn4
84 121 26 49.7 .6 1.61 2.2 1.56 2.0 .74 judge8Hamfeed8
96 121 26 49.7 .6 1.22 .8 1.16 .6 .49 judge8lamAcornl
107 121 26 49.7 .6 1.33 1.2 1.28 1.0 .63 judgeBHamAcorn2
112 119 26 49.5 .6 1.67 2.4 1.75 2.7 .51 judge13HamAcorn2
139 117 26 49.4 .6 .56 -1.6 .56 -1.6 .82 judgc6HarAcorn5
2
118
24
49
115
115
113
113
26
26
26
26
49.2
4v.2
49.1
49.1
.6 .66
.6 .59
.6 .75
:;:I :z
.b .86 -.5 .81

-.9 .76 -.9


.6 .60 -1.4 .56 -1.6
-.7
-1.2
-1.3
.91 judge3HamAcornl
.83 judge6HamAcorn3
.78 judgclHamAcor3
.73 judge5HamAcorn5
.77 judgel3HamAcornl
11 111 26 48.9
33 111 26 48.9 .6 1.59 2.1 1.63 2.3 .60 judgel2HamAcorn3
95 110 26 48.8 .6 1.05 .2 1.05 .2 .71 judge7HamAcornl
1 109 26 48.8 -6 .W .O 1.09 .3 .62 judgel3HamAcornl
74 109 26 48.8 .6 .39 -2.2 .35 -2.3 .88 judgc8Hamfccd7
15 108 26 48.7 .6 .54 -1.7 .54 -1.6 .92 judge4namAcorn2
23 108 26 48.7 .6 1.09 .3 1.14 .5 .75 judgel5HamAcorn2
56 108 26 48.7 .6 1.18 .7 1.14 .5 .74 judgel5HamAcornS
127 108 26 48.7 .6 .80 -.7 ,74 .9
~ .79 judge5HamAcor4
53 106 26 48.5 .6 1.25 .9 1.30 1.1 .71 judgelOHamAcorn5
44 105 26 48.5 .6 1.03 * 1 1.13 .5 .MI judgel3Ha#n4corn4
50 105 26 48.5 .6 .& -1.3 .64 -1.3 .75 judge6HamAcorn5
113 105 26 48.5 .6 1.25 .9 1.30 1.1 .69 judgel5HamAcorn2
............................... . ......93
...............
I... ........................
..6 .TJ judge3namfeed6
I
I 58
146
72
65
62
59
26
26
26
45.6
45.3
45.1
-.31 .81
.b
.6
.6
.58 -1.5 .55 -1.4
.55 -1.6 .50 -1.5
.81 judgelHamFeed6
.80 judge5HamFeed6

I%
I
152

71
149
59
::
51
SO
26
26
26
26
26
45.1
44.8
44.6
44.5
44.4
.6 .60 -1.4 .55 -1.4
.6 1.86
.6 1.16
3.1 1.47
.6 .97 -.l 1.00
.6 .93
.6 .94 -.2 .89
1 .4
.o
.2
.3
-
~
.77 judgeVHamFeed-5
.66 judgel2HamFeed6
.60 judge6HamFeed6
.73 judge4HamFeed6
.61 judge6HamFeedh
153 49 26 44.3 .6 1.00 .o 1.01 .o .64 judgelOHamFeedb
59 47 26 44.1 .6 .49 -1.8 .45 -1.6 .80 judge4HamFeed6
66 45 26 44.0 .6 1.18 .6 .95 -.l .65 judgel2HamFeed6
150 43 26 43.8 .6 .70 -1.0 .71 - .8 .66 judge7HamFeed6
I 148
147
42
34
26
26
43.7
42.9
.6 1.34 1.2 1.00
.7 .26 -2.4 .25 -1.9
.o .67 judge4HamFeedb
.82 judge3HamFeed6
MEAN 93. 26. 47.6 .6 1.01 . O 1.04 .2
S.D. 22. 0. 1.7 .O .43 1.6 .50 1.7
MEASURING QUALITY OF HAM BY RASCH MODEL 409

Validity is determined from the discrepancy between a particular observation


and its expectation. This identifies those individual observations with values which
contradict their use in the estimation of useful measures or calibrations.
Item function validity is determined by an analysis of the validity of the sample
or assessmentto that item. This identifies for review and revision items which may
not be working the way we intend (Table 3).

TABLE 3.
POORLY FITTING HAMS (ITEMS IN ENTRY ORDER)

This t a b l e shows the i t e m assessed by judge 11 f o r the hem 2 f o r which the Standardized o u t f i t (or
i n f i t , i f OUTFIT=N) s t a t i s t i c i s greater than the m i s f i t c r i t e r i o n (FtTPS or FlTl=). The assessmnt
codes a r e l i s t e d i n t h e i r sequence order i n the data f i l e . The residuals are standardized assessment
score residuals, uhich h i v e a rnodelled expectation of 0, and a variance of 1. Negative residuals
i n d i c a t e t h a t the Level of the observed assessmnt uas less than eapected. P o s i t i v e residuals i n d i c a t e
that t h e l e v e l o f the observed assessmnt uas m r e than expected.
i.e. The item e n t r y no. 11" Uhich correspond t o Wryness o f Lean" uas asses i n the l e v e l 2, the
corresponding residual i s - 2 uhich mans the h m acorn no.2 tested by the judge no. 11 was assessed
Less than expccted.
The item e n t r y no. 17ruhich correspond t o V i i t t e r taste" uas asses i n the l e v e l 8 , the corresponding
residual i s 5 uhich mans the ham acorn no.2 tested by the judge no. 11 uas assessed m r e than
expected

For the assessment 110, which correspond to a ham fed by acorn and assessed
by the judge 1 1, item 17 misfits the model. It has a high standardized residual of
5 when applied to the ham, (order of entry number 17 and means that it has a score
significantly higher than expected, that is, given all items and hams, the item 17 for
that ham and judge are significantlytoo large. The same can be said for the item
25, with a standardized residual of 4 (Table 3).
Ham observation validity is determined by an analysis of the validity of the
observations of that ham. This identifies review items which may not have an
observed result in the way expected (Table 4). Item 26 misfit the model, although
there are not high residuals, but there are many positives and negatives, which
means that judges have no unanimous criteria about this cellar flavor.

CONCLUSIONS

Ham quality can be considered as a latent variable defined by a set of selected


items. Applying Rasch model as an instrument for measurement ham quality, all
the scores have been summarized into a number which is the measure for hams and
items quality. The application of this model to Iberian ham sensorial analysis is a
410 C. GARCIA, J. VENTANAS, T. ANTEQUERA, J. RUIZ, R. CAVA and P. ALVAREZ

good method to determinate with objectivity the ham quality, and shows (Fig. 2 )
the higher quality of ham from acornfed pig.

0
0
0
0 0
0

A
a A

LOW QUALITY

FIG. 2. THIS MAP SHOWS ACORN HAMS FROM TABLE I1 HAVE BETTER QUALITY
THAN FEED HAMS.

These data fit the model well. Indices for suitability are good, although there
are some hams and items that misfit the model for quality according to the criteria
of 15 experienced ham tasters; it is necessary to look at the data closely in those
items and hams with high residuals in order to find an explanation for these
anomalies.
MEASURPJG QUALITY OF HAM BY RASCH MODEL 41 1

TABLE 4.
POORLY FITTING FASES (HAMS IN ENTRY ORDER)

NUMBER NAME POSITlON MEASURE INFIT CUTFIT MISFIT

26 "Cellar Ham" f lavour 1 52.7 6.7 4.9 OVER 2

76 I 1

ASSESSMENT: 0 5 0 1 5 0 7 4 0 2 4 2 6 0 0 2 2 2 0 6 0 0 4 1 0
RESIDUAL: -1 2-1 2-1 2 1-1 1 2-1-1 -1 1-1 1 -1

126

ASSESSUENT: 4 0 6 8 0 0 0 3 0 0 0 1 2 3 1 8 0 1 2 6 0 0 0 0 0
RESIDUAL: 1-1 2 2-1-1 -1 -1 2 2

This t a b l e shows the assessments of item 26, from every judges and every hams, for uhich the
Standardized o u t f i t ( o r i n f i t , i f W T F I T = N ) s t a r i s t i c i s greater than the m i s f i t c r i t e r i o n (FITP- or
FITII). The assessment codes a r e l i s t e d i n t h e i r sequence order i n the data f i l e . The residuals are
standardized assessment score residuals, uhich have a modelled expectation of 0, and a variance of
1. NegafIve residuals i n d i c a t e that the observed assessment n05 less correct than expected.

Fit analysis is a good way to find out how the items and hams work, and to
identify items and hams with those causes that bring about high residuals through
unexpected scores.
This methodology can be applied to any kind of ham and different sets of items
and different judges.

REFERENCES

ALVAREZ, P. 1992. Medida de la calidad de 10s vinos de Tierra de Barros y


calibraci6n de las caracteristicas sensoriales. Actas de las V jornadas
Latinoamericanas de Viticultura y Enologfa. Montevideo. Uruguay.
ALVAREZ, P., MORAN, J.C. and WRIGHT, B.D. 1993. Quality of Life. VII
Objective Measurement Workshop, Emory University Press, Atlanta, Georgia.
ANDRICH, D. 1988. Rasch Model for Measurement, Murdoch University, Sage,
London.
GUEDES, P., BERTRAND, A. and ALVAREZ, P. 1995 Wine Characterization by
412 C. GARCIA, J. VENTANAS, T. ANTEQUERA, J . RUIZ, R. CAVA and P. ALVAREZ

Multivariates Statistical Analysis of the Sensory and Chemical Data. Trends in


Flavour Research (In press).
HORIMOTO, Y., DURANCE, T., NAKAI, S. and LUKOW. 0. M. 1995. Neural
networks vs principal component regression for prediction of wheat flour loaf
volume in baking tests. J. Food Sci. 60,429-433.
LOZANO, M.M. 1993. Estudio de mieles monoflorales de Extremadura. Theses.
University of Extremadura. Spain.
RASCH, G. 1980. Probahilistic Models for Some Intelligence and Attainment Test.
The University of Chicago Press.
VENTANAS, J., CORDOBA, J.J., ANTEQUERA, T., GARCIA, C.,
LOPEZ-BOTE, C. and ASENSIO, M.C. 1992. Hydrolysis and Maillard
Reactions During Ripening of Iberian Ham. J. Food Sci. 57, 813-815.
WRIGHT, B.D. and DOUGLAS, G.A. 1977. Best Procedures for Sample Free Item
Analysis. App. Psych. Measure. /, 281-294.
WRIGHT, B.D. and LINACRE, J.M. 1991. Bigsteps, MESA Press, Chicago,
Illinois.
WRIGHT, B.D. and MASTERS, J. 1982. Rating Scale Analysis. MESA Press,
Chicago, Illinois.
WRIGHT, B.D. and MEAD, R.J. 1976. Calibrating Items with the Rasch Model.
In Research Memorandum, Statistical Laboratory Department of Education.
Vol 23. University of Chicago Press.
WRIGHT, B.D. and STONE, M.H. 1978. Best Test Design, MESA Press, Chicago,
Illinois.

You might also like