Professional Documents
Culture Documents
Full download Easy Statistics for Food Science with R Abdulraheem Alqaraghuli file pdf all chapter on 2024
Full download Easy Statistics for Food Science with R Abdulraheem Alqaraghuli file pdf all chapter on 2024
Abdulraheem Alqaraghuli
Visit to download the full and correct content document:
https://ebookmass.com/product/easy-statistics-for-food-science-with-r-abdulraheem-a
lqaraghuli/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
https://ebookmass.com/product/applied-statistics-for-
environmental-science-with-r-1st-edition-abbas-f-m-alkarkhi/
https://ebookmass.com/product/applied-statistics-with-r-a-
practical-guide-for-the-life-sciences-justin-c-touchon/
https://ebookmass.com/product/basic-statistics-with-r-reaching-
decisions-with-data-stephen-c-loftus/
https://ebookmass.com/product/the-new-statistics-with-r-an-
introduction-for-biologists-2nd-edition-andy-hector/
Applied Statistics: Theory and Problem Solutions with R
Dieter Rasch Rostock
https://ebookmass.com/product/applied-statistics-theory-and-
problem-solutions-with-r-dieter-rasch-rostock/
https://ebookmass.com/product/statistics-for-business-
economics-14e-edition-david-r-anderson/
https://ebookmass.com/product/r-programming-for-actuarial-
science-peter-mcquire/
https://ebookmass.com/product/statistics-for-ecologists-using-r-
and-excel-data-collection-exploration/
Abbas
To the memory of my parents (deceased)
To my children Atheer, Hibah, and Farah
Wasin
To the memory of my father (deceased)
To my mother
This page intentionally left blank
Contents
vii
viii Contents
Easy statistics for food science with R software was written in a simple way to introduce some statistical methods that
are important to graduate students, postgraduate students, and researchers who work in the food science or food engi-
neering fields with applications in food science. Specifically, this book was written to help researchers from different
fields to analyze their data and make valid decisions. The interpretation of the results is carried out in a step-by-step
manner and in an easy and clear style to enable nonstatisticians to understand and use it in their research. This book is
not comprehensive because the topics were selected based on what is commonly used in the field.
The analysis is carried out step by step, and the interpretation of the results is given for each example by matching
the results to the area of study where the data were obtained. The book focuses on the applications of univariate and
multivariate statistical methods in the field of food science. We use real data obtained from research in the School of
Industrial Technology, USM over more than 10 years of work in food science.
The development of modern statistical packages makes the analysis of data easier than before. Thus, no mathemati-
cal proofs are given in this book. Instead, the focus is on the application of statistics and correct methods for the analy-
sis and interpretation of data.
R statistical software is used throughout the book to analyze the data. The beauty of R is that it is open source, and
you can label your work based on your ideas and arrangement to produce beautiful graphs with high resolution. R is
available over the internet under the General Public License (GPL online) for the Windows, Macintosh, and Linux oper-
ating systems.
Finally, we wish to thank our colleagues and friends for their continuous support, especially Dr. Yusri Yusup for his
valuable comments. We would like to extend our thanks to the R software community and R family (R users and contri-
butors to R). This book would not have been possible without the information provided online, which is easy to obtain.
We thank the University of Kuala Lumpur (UnikL-MICET) for its support.
ix
This page intentionally left blank
Chapter 1
Introduction
Learning outcomes
At the end of this chapter, you should be able
G To describe multivariate analysis.
G To understand the benefits of multivariate methods.
G To arrange the results of multivariate data to be ready for analysis.
G To understand the difference between univariate and multivariate concepts.
G To know how and where to use multivariate data.
G To comprehend the univariate normal distribution.
G To comprehend the multivariate normal distribution.
^ ^ ^ ... ^ ... ^
^ ^ ^ ... ^ ... ^
Group pH TSS L* a* b* WHC40 WHC60 WHC80 OHC40 OHC60 OHC80 Viscosity Texture
Gpe 4.64 1.83 45.16 5.28 21.01 5.15 5.21 5.70 0.78 0.68 1.10 60.07 32.70
Gpe 4.32 1.73 43.10 6.03 23.02 5.16 4.87 5.97 0.78 0.77 1.06 46.90 34.16
Gpe 4.60 1.77 44.72 6.42 23.38 5.03 5.14 5.65 0.79 0.75 1.17 55.70 33.88
Gpe 4.59 1.57 37.82 5.89 21.35 5.15 4.95 6.16 0.76 0.75 1.02 49.97 37.06
Gpe 4.55 1.53 38.64 5.67 24.30 5.20 5.37 6.50 0.74 0.78 0.95 58.83 37.26
Gpe 4.33 1.67 39.85 5.36 23.73 4.99 5.53 6.00 0.80 0.77 1.02 57.53 38.02
Gpe 4.30 1.60 37.04 5.52 22.88 4.58 5.36 6.21 0.69 0.76 1.05 57.70 38.89
Gpe 5.08 1.70 36.35 4.94 21.81 5.20 5.85 5.97 0.75 0.80 1.01 46.73 39.20
Gpe 5.31 1.80 48.73 3.79 26.58 4.34 5.53 5.15 0.73 0.71 1.03 55.03 38.80
Gpe 5.33 1.80 38.42 4.38 22.66 4.97 4.81 5.71 0.85 0.75 1.02 54.53 36.90
Gpe 5.24 1.83 45.93 4.26 27.07 4.14 5.27 5.88 0.76 0.78 0.96 52.97 39.68
Gpe 5.26 1.90 34.83 4.89 21.45 5.03 4.82 5.64 0.72 0.78 0.97 54.87 40.91
Rpe 4.86 3.30 32.43 5.26 11.16 5.39 6.43 6.65 0.94 0.94 1.30 66.90 54.19
Rpe 5.31 3.20 35.61 4.77 13.10 6.10 6.56 8.10 1.06 0.98 1.33 73.47 43.99
Rpe 5.32 3.60 35.25 5.29 11.55 5.55 6.34 8.19 0.97 1.04 1.28 77.23 35.94
Rpe 5.31 3.53 32.50 5.52 11.02 6.19 6.64 8.10 0.97 0.92 1.07 79.50 55.17
Rpe 5.45 3.23 39.85 5.53 12.57 6.34 6.58 8.48 0.90 1.01 1.24 80.20 42.62
Rpe 5.69 3.47 40.31 5.55 13.07 6.21 6.72 9.26 1.01 0.96 1.39 83.90 42.62
Rpe 5.61 3.53 38.14 5.74 12.49 6.21 6.47 9.26 0.89 0.92 1.26 76.27 45.43
Rpe 5.65 3.57 39.29 5.98 12.79 6.16 6.23 8.29 1.00 0.97 1.30 79.40 47.50
Rpe 5.65 3.47 36.68 5.72 12.08 5.94 6.47 8.31 0.78 1.00 1.21 82.90 59.92
Rpe 5.68 3.50 41.08 5.32 12.97 6.17 5.90 7.92 0.96 0.98 1.37 78.30 63.24
Rpe 5.58 3.53 39.95 5.57 12.81 6.55 6.20 7.84 0.90 1.02 1.29 66.80 57.61
Rpe 5.47 3.63 40.37 6.34 14.01 6.34 5.59 7.88 0.81 1.05 1.31 72.73 59.93
Gpu 4.49 1.07 70.97 3.15 15.57 3.83 5.56 6.07 0.91 0.54 0.87 35.07 0.54
Gpu 4.37 1.17 71.70 3.41 15.22 3.72 5.52 6.03 0.91 0.57 0.83 37.00 0.63
(Continued )
TABLE 1.2 (Continued)
Group pH TSS L* a* b* WHC40 WHC60 WHC80 OHC40 OHC60 OHC80 Viscosity Texture
Gpu 4.75 1.20 68.23 3.67 15.09 4.07 5.37 6.15 0.87 0.54 0.86 40.13 0.66
Gpu 5.64 1.03 64.37 3.53 14.69 4.01 5.52 6.24 0.81 0.59 0.84 38.60 0.72
Gpu 4.73 1.17 74.10 2.83 15.34 4.03 5.60 6.35 0.84 0.42 0.97 42.27 0.72
Gpu 4.59 1.13 79.25 2.49 17.69 4.08 5.78 6.53 0.77 0.47 0.85 39.13 0.70
Gpu 4.54 1.23 79.15 2.54 17.46 3.93 5.80 6.43 0.79 0.53 0.88 38.73 0.71
Gpu 5.37 1.23 78.70 1.57 16.77 3.89 5.61 6.21 0.89 0.42 0.87 40.20 0.81
Gpu 5.58 1.27 74.78 1.86 19.54 3.93 5.58 6.43 0.71 0.43 0.71 42.53 0.70
Gpu 5.65 1.37 76.62 1.68 19.29 3.74 5.79 6.43 0.73 0.45 0.79 44.67 0.68
Gpu 5.53 1.30 77.08 1.66 20.04 4.02 5.74 6.46 0.71 0.45 0.85 47.47 0.58
Gpu 5.47 1.43 75.19 1.99 21.69 3.97 5.99 6.44 0.64 0.64 0.81 45.53 0.63
Rpu 5.60 4.33 74.39 2.52 11.51 1.53 1.61 4.86 0.81 0.82 1.15 84.13 2.36
Rpu 5.57 4.03 72.00 2.88 11.85 1.50 1.74 4.04 0.82 0.84 1.03 85.17 2.49
Rpu 4.76 3.77 67.38 5.07 20.23 1.52 1.81 4.98 0.77 0.81 1.13 91.67 2.46
Rpu 5.56 4.10 70.00 3.13 12.07 1.69 1.76 5.03 0.79 0.83 1.15 87.40 2.72
Rpu 4.95 3.97 68.14 4.10 16.71 1.42 1.77 4.04 0.80 0.83 0.99 87.23 2.39
Rpu 4.84 4.40 69.82 3.67 16.11 1.28 1.56 5.01 0.69 0.87 0.94 87.33 2.39
Rpu 5.07 4.50 70.82 2.70 12.40 1.25 1.68 5.01 0.79 0.85 1.05 86.90 2.23
Rpu 4.94 4.43 72.20 2.67 12.96 1.08 1.68 4.84 0.87 0.78 1.00 90.00 2.29
Rpu 5.04 4.30 67.12 3.87 15.93 1.54 1.88 4.83 0.79 0.83 1.11 86.97 2.08
Rpu 4.93 4.57 72.69 2.59 13.31 1.18 1.58 3.99 0.70 0.73 0.95 89.73 2.20
Rpu 5.05 4.30 70.82 3.00 13.71 1.26 1.78 4.77 0.80 0.83 1.07 90.27 1.97
Rpu 5.21 4.40 74.86 2.42 12.97 1.24 1.63 4.59 0.83 0.80 1.04 87.70 2.30
Introduction Chapter | 1 5
TABLE 1.3 The Arsenic and Heavy Metal Contents in Cockles Obtained From 20 Sites at the Juru and Jejawi Rivers
(mg/L)
Location Cr As Cd Zn Cu Pb Hg
Juru 0.20 2.38 0.82 0.23 0.20 0.13 1.40
(Continued )
6 Easy Statistics for Food Science with R
Location Cr As Cd Zn Cu Pb Hg
Jejawi 0.18 2.88 0.86 0.19 0.17 0.12 1.21
fatty acid (PUFA), monounsaturated fatty acid (MUFA), oil color values (lightness (L*) value, redness (a*), and yellow-
ness (b*)), and moisture content (%). The data are given in Table 1.4.
Observe that the data do not provide useful information for making a comparison between different brands based on
the selected parameters. The objective of the study was to assess the quality characteristics of tapioca chips regarding
their fat oxidation, fat content, fatty acid composition, and moisture and oil color values. Furthermore, the objectives
include finding the similarities between different brands based on the selected parameters and finding the source of the
differences between these brands. The last goal was to identify the parameters responsible for the distinctions between
the brands.
Example 1.4: Antioxidant in dates—Edible parts of date palm (Phoenixdactylifera) fruits (DPF) were sampled and
analyzed for their antioxidant activities (AA) using the Trolox equivalent antioxidant capacity (TEAC) method, 2,2’-
azinobis (3-ethylbenzothiazoline-6-sulfonic acid) radical cation (ABTS1) assays and the ferric reducing/antioxidant
power method (FRAP assay). The total flavonoid content (TFC) and total phenolic content (TPC) of the DPF were mea-
sured using the aluminum chloride colorimetric and FolinCiocalteaumethods, respectively. Four types of soft dates
(SD), namely, Jiroft dates, Honey dates, Kabkab dates, and Bam dates, were used; three types of semidry dates (SDD),
namely, Piarom dates, Zahedi dates, and Sahroon dates, and one type of dry date (DD) (Kharak dates) were also used.
The data are given in Table 1.5.
The main goal of this study was to evaluate the AA of methanolic extracts from eight different types of DPF using
the ABTS and FRAP methods. We investigated the similarities between different types of dates to study the relation-
ships between various parameters. The goal was to understand the behavior of each variable in the presence of the other
variables. Furthermore, we tested the differences among the selected varieties of dates regarding the antioxidant activi-
ties (ABTS and FRAP assay) and antioxidative compounds (flavonoids content and total phenolic). The final objective
was to identify the source of the variation in the dates and to detect the effects of each parameter on the degree of dif-
ferentiation among the same varieties of dates.
Brand Fat Content Moisture L* a* b* Peroxide p-Anisidine Acid Value TBA SAF MUFA PUFA
T1 20.30 2.29 73.01 14.65 83.59 4.63 13.86 4.38 26.60 50.03 40.07 9.90
T1 19.96 2.24 73.00 14.63 83.55 3.45 13.91 4.48 28.10 50.38 39.82 9.80
T1 20.29 2.27 73.00 14.65 83.56 4.22 13.86 3.96 26.40 50.34 39.84 9.83
T2 18.57 3.29 23.77 44.29 40.76 17.05 10.09 24.41 79.90 49.33 27.78 8.53
T2 18.73 3.41 23.76 44.28 40.73 17.12 10.09 22.32 79.70 49.31 27.57 8.54
T2 18.44 3.33 23.75 44.28 40.70 16.97 10.01 22.20 81.10 49.81 28.10 8.47
T3 17.74 3.74 79.40 12.26 48.82 2.70 6.74 4.05 10.00 51.25 38.76 9.99
T3 17.22 3.75 79.42 12.26 48.84 2.66 6.73 3.96 9.97 51.63 38.46 9.92
T3 17.46 3.24 79.48 12.29 48.82 2.59 6.74 4.21 8.90 51.78 38.33 9.89
T4 26.05 5.32 70.05 11.45 77.16 10.73 8.53 7.97 62.40 50.77 39.33 9.90
T4 26.42 5.47 70.05 11.44 77.16 10.74 8.53 8.24 62.40 51.06 39.13 9.81
T4 26.23 5.39 70.04 11.43 77.16 10.33 8.53 8.56 57.00 50.96 39.27 9.78
T5 20.62 2.79 82.34 12.65 75.25 8.67 11.97 4.47 39.70 52.19 37.69 10.12
T5 20.89 2.62 82.34 12.63 75.20 8.55 11.96 4.20 37.80 52.33 37.72 10.05
T5 20.51 2.57 82.34 12.65 75.14 8.71 11.97 4.46 43.20 52.23 37.72 10.05
T6 29.91 4.57 67.27 41.11 114.58 10.94 2.53 3.42 7.44 51.48 37.98 10.54
T6 30.21 4.60 67.25 41.08 114.55 10.76 2.45 2.99 4.99 51.63 37.89 10.48
T6 29.92 4.55 67.24 41.08 114.57 10.98 2.65 2.91 5.11 51.70 37.87 10.43
T7 25.99 4.55 72.86 18.61 99.22 3.60 14.24 9.93 4.99 48.61 30.17 9.30
T7 26.04 4.47 72.85 18.62 99.26 2.70 15.08 8.51 4.68 49.02 30.18 9.25
T7 25.60 4.47 72.83 18.61 99.23 4.15 15.10 8.31 4.52 49.48 30.28 9.18
8 Easy Statistics for Food Science with R
The distribution is a bell-shaped curve, as illustrated in Fig. 1.1. Fig. 1.1 is generated using R statistical software.
The codes for generating normal distribution curve are given in the Appendix.
b. Multivariate normal distribution P
The multivariate normal density with mean vector μ and covariance matrix is given in (1.2).
1 0 P 21
2ðY2μÞ ðY2μÞ=2
gð Y Þ 5 e
k=2 P1=2
(1.2)
ð2πÞ
0P
Where k is the number of variables and ðY2μÞ P21 ðY 2 μÞ is the Mahalanobis distance (statistical distance). We
denote this k-dimensional normal density by Nk ðμ; Þ.
Note
G Use an appropriate transformation of the data if one or more variables under study violate the normality assumption,
as when the data are highly skewed, with several extreme values (high or low), called outliers, or repeated values.
Introduction Chapter | 1 9
G If all the individual variables are normally distributed, then it is assumed that the joint distribution is a multivariate
normal distribution.
G The real data never follow a multivariate normal distribution exactly, but the normal density is used as an approxi-
mation of the true population distribution.
FURTHER READING
Abbas, F. M. A., Foroogh, B., Liong, M. T., & Azhar, M. E. (2008). Multivariate statistical analysis of antioxidants in dates (Phoenix dactylifera).
International Food Research Journal, 15, 193200.
Alkarkhi, A. A. F. M., Ismail, N., & Easa, A. M. (2008). Assessment of arsenic and heavy metal contents in cockles (Anadara granosa) using multi-
variate statistical techniques. Journal of Hazardous Materials, 150, 783789.
Alkarkhi, A. F. M., Saifullah, B. R., Yeoh, S. Y., & Azhar, M. E. (2011). Comparing physicochemical properties of banana pulp and peel flours pre-
pared from green and ripe fruits. Food Chemistry, 129, 312318.
Alqaraghuli, W. A. A., Syuzeliana, S., Yusri, Y., & Alkarkhi, A. F. M. A. An investigation of oil stability indices of fried tapioca chips in the market:
Multivariate analysis.
Blogger. R graph gallery: A collection [Online]. Available: ,http://rgraphgallery.blogspot.my/2013/04/shaded-normal-curve.html. Accessed
07.07.15.
Bryan, F. J. M. (1991). Multivariate statistical methods : A primer. Great Britain: Chapman & Hall.
Daniel, H. 2013. Blog archives, high resolution figures in R [Online]. Available: ,https://www.r-bloggers.com/author/daniel-hocking/. Accessed
15.07.17.
Johnson, R. A., & Wichern, D. W. (2002). Applied multivariate statistical analysis. New Jersey: Prentice Hall.
Rencher, A. C. (2002). Methods of multivariate analysis. New York: J. Wiley.
Another random document with
no related content on Scribd:
riconosciuta probità [725], avea la ferma persuasione che l'essere
severi ne' giudizî era ancora una delle poche cose, che potessero
fare argine al rifluire della parte popolare [726]. Q. Titinio, di una
famiglia plebea, che avea dato alla repubblica tribuni, i quali si erano
opposti in altro tempo (193 av. C.) al trionfo di Q. Metello, e cavalieri,
che avevano lottato per le prerogative dell'ordine [727], dovea ora
vedere in Verre non solo l'aderente di Silla e de' Metelli, ma anche il
pretore, da cui il suo fratellastro C. Junio era stato, o pretendeva di
essere stato danneggiato [728]. C. Marcello, discendente
dell'espugnatore di Syracusae, era anche stato pro-pretore in Sicilia
(79 av. C.) [729] e, mentre da un lato era legato da un vincolo di
patronato ereditario a' Siciliani [730], dall'altro era imparentato con la
famiglia Junia [731].
Di tutti questi, che conosciamo, se non il solo favorevole, certamente
il più favorevole era M. Metello, per i rapporti di famiglia, per l'aiuto
da lui avuto nelle elezioni, per tradizione politica.
Cicerone ne dovè ricusare degli altri ben più compromessi, se si
adattò a ritenere M. Metello; tra gli altri dovette ricusare quel M.
Lucrezio, che rimproverava quasi a Verre di aver voluto ritenere e
che non sappiamo in quali rapporti fosse con Verre [732].
Tuttavia, o che guardasse agli altri giudici -- ve ne doveano ben
essere molti altri; nel processo di A. Cluenzio erano trentadue [733] --
o che, chiuso in quelle distrette, prendesse a fare quell'alchimia, che
gli accusati sogliono fare, almanaccando su tutti i possibili rapporti
con i giudici [734]; cominciò forse a riaversi un cotal poco da
quell'abbattimento, che Cicerone gli appone. Che quando, con
l'intervallo di pochi giorni, l'elezioni ebbero luogo e dettero la vittoria
alla sua parte, ogni abbattimento era dileguato, ed egli credeva
vedere in esse l'auspicio e l'augurio della vittoria.
Le elezioni.