301181 Food Evaluation

Investigation of the effect of fat type on the quality and shelf life of biscuits

This laboratory exercise aims to conduct a sensory training to develop the necessary skills to be able to characterize baked foods (especially) cereals according to their sensory qualities. The sensory skills developed will then be applied to investigate the shelf life of biscuits manufactured using various fat types to establish how the biscuit quality and shelf life are affected by the type of  fat used in the product. The different fat types to be investigated include vegetable oils, butter and coconut oil.   

Sensory analysis is a scientific procedure which can be applied to interpret the reaction of a particular subject to characteristics of substances such as food due to stimulation by smell, taste or touch (Obaroakpo, Iwanegbe, & Ojokoh 2017). In the case of food substances such as biscuits, the taste, smell and touch form the basis of sensory analysis. Obaroakpo, Iwanegbe, & Ojokoh (2017) characterized sensory evaluation as an integral part of product and quality control. According to Civille & Oftedal (2012), sensory analysis can be applied to study the shelf life of food substances and various sensory tests are applied in food-science research. Sensory analysis is a quantitative approach which involves the gathering of numerical data to establish the connection between human perception and the properties of products. The data produced from the use of human observers contains many variables and a sensory panel of trained inspectors for data generation is highly heterogeneous. Once data is gathered, appropriate methods of statistical analysis are used to establish whether the observed association between the properties of the product and the sensory responses represent actual unbiased responses and not just from uncontrolled variations (Chopra, Dhillon, Rani, & Singh, 2018). Thus sensory analysis should be designed to be precise and accurate to the highest degree possible to avoid false positive results.

A major ingredient of baked foods, fat gives baked foods important sensory characteristics. These include texture, odour, taste and colour. All these factors determine the level of consumer acceptance for a particular product. The shelf life of any consumer product is critical and is of interest to all the parties involved in the food chain, especially the producer and the consumer. The evaluation of the shelf-life of baked products such as biscuits incorporates well designed acceptability tests which involve sensory analysis. Food contents such as moisture and water vapour have a major impact on the shelf life of biscuits (Rigo et al., 2014).

According to Gebreselassie & Clifford (2016), fat is the principal component that gives biscuits their overall texture. Oxidation of fats is a major cause of the deterioration of food substances with high levels of fat. According to Caruso et al., (2016), the oxidation of lipids results in off-odour and off-flavour that compromise of the quality and acceptability of products. Besides, this effect affects the nutritional content of the food (Devi & Khatkar 2016). The modern food industry is driven by the goal to deliver satisfactory products to the consumers for a long time. Another major objective is to offer high sensory quality levels which is associated with fresh products (Mieszkowska & Marzec 2015). Baked foods such as biscuits exhibit a negative change during storage which involves the formation of unwanted substances such as aromatic compounds. According to Thomas (2018), the shelf life of a product can be defined as the time period during which small variations occur in a product without significantly lowering the acceptability and safety. According to Wirkowska-Wojdy?a, Bry?, Górska, & Ostrowska-Lig?za (2016), the low humidity in products such as cookies coupled with high fat content makes these products highly susceptible to fat oxidation.

Texture is a very important property of biscuits. The texture on the other hand is dependent on the cooking formula and cooking regime. The primary ingredients of biscuits include fat, sugar and soft wheat flour. Other minor ingredients such as milk, salt and flavouring agents are also present. The moisture content of biscuits is considerably low (about 1 to 5 %) (Rigo et al., 2014).  Different types of fat have different effects on the texture properties of biscuits. Fats can be sourced from animals or from vegetables and each type has distinct effects on the sensory properties of biscuits (Joseph 2016). That level of fat in a product also determines the sensory characteristics of the product. High fat levels raise the tenderness of cookies while low fat concentration produces biscuits of high breaking strength. Fat plays a critical role in the mechanical properties of baked products since it significantly reduces the size of the air cells in the structure greatly lowering the fracture stress of the product (Adeola & Ohizua 2018). Oils and fats influence the integrity of the structure of a product through interaction with other non-fat ingredients. Devi & Khatkar (2017) studied how oils and fats influence the shelf life and structural integrity of the final product. Fats and oils exhibit greater variations in composition compared to other biscuit ingredients. This is partly due to the fact that they are obtained from a large number of different plants and animals. The physiochemical properties of oils and fats are rather complicated. Each fat and oil type has a range of chemical, physical and compositional parameters (Gebreselassie & Clifford 2016).   

Fats have the function of dough shortening. This means that they have the ability to weaken and lubricate the structure of food products to yield a desired texture. Devi & Khatkar (2016) investigated the effect of fat types and fat quantity on the quality of produced biscuits. Their study revealed that biscuit thickness was considerably higher for biscuits produced using hydrogenated oil or fat while better surface properties and higher crispness was obtained with biscuits made through bakery shortening. Mamat & Hill (2012) investigated the effect of different fat types on the quality of cookies and the dough. They established that, the use of liquid oil in biscuits gave the final product a harder texture in contrast to hydrogenated and bakery fat. When considering the functionality of fats in the dough, the ratio of solid fat to the total fat is a crucial factor. This ratio is known as the solid fat index (SFI). The texture and functional performance of fats and products containing fat is determined by the SFI. Bakery fats such as butter and margarine

The sensory analysis was performed through qualitative description evaluation of the properties of the biscuits including colour, crispness, saltiness, sweetness, hardness, aroma, and acceptability.

Sensory panel trainings

A series of training exercises were carried out to familiarise with the connection between the intensity of the sensory attributes and the category rating.

Color

The sensory ability to associate color changes with certain properties of biscuits was inculcated through a review of a range of colours between extremely pale and extremely dark on a 15 cm rating scale. Some of the categories identified between the two extremes include slight, moderate, very slight, moderate-strong and threshold.

Taste

The sensory training for sweetness involved familiarisation with attributes of biscuits such as sweetness, saltiness, and sourness. Sensory training for sweetness included tasting a range of solutions between two extremes, not sweet and strongly sweet. Several solutions with different concentrations in terms of the mass of sugar dissolved in a litre of water were prepared. Each concentration was assigned a unique rating on the 15 cm line scale. A similar procedure was used to establish similar ratings for saltiness using solutions of different concentrations.

To establish the sensory scale for sourness, solutions with different concentrations of citric acid were used. The two extremes in this case were identified as the taste of a solution being sour and another solution of different concentration being strongly sour. Intermediate sensory attributes were given appropriate ratings. The sensory of bitterness was measured between two extremes ranging from not bitter to strongly bitter. Attributes between the two extremes were given appropriate ratings on the 15 cm scale.

Texture

The hardness of a food product was rated on a scale between two extremes ranging from extremely soft to extremely hard. The hardness of semisolid foods was measured through compression of a sample of the product between the tongue and the palate.

This part of the laboratory exercise involved applying the sensory skills generated during the panel training to characterize biscuits produced using different types of fats and to establish how these affect the quality and shelf life of the biscuits.

In this exercise, three different types of fats/oils were used. These include,

1. Vegetable shortening
2. Unsalted butter

• Coconut oil

Three groups were formed each handling the production of biscuits using one of the above fat/oil types. The given formulation for the production of the biscuits in which the given proportions for the biscuit ingredients were used. The dough developed after mixing the ingredients in the appropriate proportions was then cut into portions of about 4.2 cm and baked in an oven a temperature of about  for about 10 minutes according to the directions given in the manual. The biscuits were then allowed to cool on trays.

Average ratings and standard deviation

 

The figure above shows the average ratings and the standard deviations for the various sensory characteristics of biscuits.

ANOVA with Turkey Post Hoc test

Colour

Table 1: Analysis of variance for biscuit colour

SUMMARY
Groups	Count	Sum	Average	Variance
163	11	71.7	6.518182	9.163636
798	11	42.6	3.872727	5.670182
502	11	69.6	6.327273	6.588182
987	11	70.2	6.381818	5.451636
314	11	55.8	5.072727	2.738182
756	11	78.3	7.118182	11.89764
ANOVA
Source of Variation	SS	df	MS	F
Between Groups	77.80364	5	15.56073	2.249231
Within Groups	415.0945	60	6.918242
Total	492.8982	65

 
Since the critical F value is greater than our F statistic, we conclude that there is no significant difference in the means. This means that there is no difference between the average colour ratings for week 0 and week 2.

Table 2: Analysis of variance for biscuit aroma

SUMMARY
Groups	Count	Sum	Average	Variance
163	11	73.3	6.663636	10.63455
798	11	80.9	7.354545	10.40473
502	11	91.5	8.318182	11.29364
987	11	58.3	5.3	10.11
314	11	73.9	6.718182	8.403636
756	11	99.4	9.036364	7.276545
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	96.47227	5	19.29445	1.991751	0.092788	2.36827
Within Groups	581.2309	60	9.687182
Total	677.7032	65

 

Since the critical F value is greater than our F statistic, we conclude that there is no significant difference in the means. This means that there is no change in the aroma of the biscuits for week 0 and week 2.

Table 3: Analysis of variance for rancidity aroma

Anova: Single Factor
SUMMARY
Groups	Count	Sum	Average	Variance
163	11	17.7	1.609091	2.786909
798	11	12.9	1.172727	1.304182
502	11	19	1.727273	2.646182
987	11	52.8	4.8	11.36
314	11	57	5.181818	16.83164
756	11	43.35	3.940909	10.90641
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	173.2274621	5	34.64549	4.535214	0.001423	2.36827
Within Groups	458.3531818	60	7.63922
Total	631.5806439	65

 

Since we have rejected the null hypothesis, we proceed to perform the Turkey Post Hoc test

Table 4: Turkey Post Hoc test for rancidity aroma

comparison	Abs difference	T
1 to 2	0.436363636	3.47	No significant difference
1 to 3	0.118181818	3.47	No significant difference
1 to 4	4.8	3.47	Means significantly different
1 to 5	3.572727273	3.47	Means significantly different
1 to 6	2.331818182	3.47	No significant difference
2 to 3	0.554545455	3.47	No significant difference
2 to 4	3.627272727	3.47	Means significantly different
2 to 5	4.009090909	3.47	Means significantly different
2 to 6	2.768181818	3.47	No significant difference
3 to 4	3.072727273	3.47	No significant difference
3 to 5	3.454545455	3.47	No significant difference
3 to 6	2.213636364	3.47	No significant difference
4 to 5	0.381818182	3.47	No significant difference
4 to 6	0.859090909	3.47	No significant difference
5 to 6	1.240909091	3.47	No significant difference

 

Table 5: Analysis of variance for hardness

Anova: Single Factor
SUMMARY
Groups	Count	Sum	Average	Variance
163	11	102.9	9.354545	2.592727
798	11	87.8	7.981818	5.649636
502	11	113.9	10.35455	2.420727
987	11	79.9	7.263636	5.636545
314	11	73.1	6.645455	7.398727
756	11	101.5	9.227273	6.816182
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	109.2298	5	21.84597	4.295519	0.002092	2.36827
Within Groups	305.1455	60	5.085758
Total	414.3753	65

 

Since we have rejected the null hypothesis, we proceed to perform the Turkey Post Hoc test

 

Table 6: Turkey Post Hoc test for hardness

comparison	Abs difference	T
1 to 2	1.372727	2.83	No significant difference
1 to 3	1	2.83	No significant difference
1 to 4	2.090909	2.83	No significant difference
1 to 5	2.709091	2.83	No significant difference
1 to 6	0.127273	2.83	No significant difference
2 to 3	2.372727	2.83	No significant difference
2 to 4	0.718182	2.83	No significant difference
2 to 5	1.336364	2.83	No significant difference
2 to 6	1.245455	2.83	No significant difference
3 to 4	3.215818	2.83	Means significantly different
3 to 5	3.709091	2.83	Means significantly different
3 to 6	1.127273	2.83	No significant difference
4 to 5	0.618182	2.83	No significant difference
4 to 6	1.179636	2.83	No significant difference
5 to 6	2.581818	2.83	No significant difference

Table 7: Analysis of variance for crispness

SUMMARY
Groups	Count	Sum	Average	Variance
163	11	99.8	9.072727	5.460182
798	11	99.3	9.027273	7.872182
502	11	113.7	10.33636	6.888545
987	11	94.9	8.627273	5.042182
314	11	93.2	8.472727	4.590182
756	11	114.6	10.41818	5.589636
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	39.41712	5	7.883424	1.334556	0.262064	2.36827
Within Groups	354.4291	60	5.907152
Total	393.8462	65

Since the value of p is higher than the significance level of 0.05, we cannot reject the null hypothesis hence the crispness for the three cookie types is the same and does not change significantly over the period of 2 weeks is the same.

Table 8: Analysis of variance for sweetness

SUMMARY
Groups	Count	Sum	Average	Variance
163	11	85.1	7.736364	7.010545
798	11	90.3	8.209091	5.604909
502	11	82.3	7.481818	7.375636
987	11	70.4	6.4	6.438
314	11	90.75	8.25	8.5375
756	11	80.6	7.327273	8.418182
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	25.67473	5	5.134947	0.71015	0.618149	2.36827
Within Groups	433.8477	60	7.230795
Total	459.5225	65

 

Since the value of p is higher than the significance level of 0.05, we cannot reject the null hypothesis hence the sweetness of the three cookie types is the same and does not change significantly over the period of 2 weeks is the same.

Table 9: Analysis of variance for vanilla flavour

SUMMARY
Groups	Count	Sum	Average	Variance
163	11	57.1	5.190909	11.64291
798	11	68.4	6.218182	12.05764
502	11	65.4	5.945455	9.966727
987	11	63.2	5.745455	8.764727
314	11	71.6	6.509091	14.36091
756	11	66.8	6.072727	10.43818
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	11.19348	5	2.238697	0.199791	0.961328	2.36827
Within Groups	672.3109	60	11.20518
Total	683.5044	65

Since the value of p is higher than the significance level of 0.05, we cannot reject the null hypothesis hence the vanilla flavour for the three cookie types is the same and does not change significantly over the period of 2 weeks is the same.

Table 10: Analysis of variance for acceptability

SUMMARY
Groups	Count	Sum	Average	Variance
163	11	117.9	10.71818	3.575636
798	11	117.1	10.64545	5.120727
502	11	108.8	9.890909	5.358909
987	11	104.5	9.5	3.082
314	11	95.4	8.672727	2.792182
756	11	112.7	10.24545	6.554727
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	32.98182	5	6.596364	1.494408	0.20515	2.36827
Within Groups	264.8418	60	4.41403

 
Since the value of p is higher than the significance level of 0.05, we cannot reject the null hypothesis hence the acceptability for the three cookie types over the period of 2 weeks is the same.

Colour

Table 11: Paired t-test for colour

t-Test: Paired Two Sample for Means
	week 0	week 2
	163	987	798	314	502	756
Mean	6.52	6.381818	3.873	5.07	6.33	7.12
Variance	9.16	5.451636	5.67	2.74	6.59	11.9
Observations	11	11	11	11	11	11
Pearson Correlation	0.27		0.272		0.6
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	0.14		-1.59		-0.93
P(T<=t) one-tail	0.45		0.071		0.19
t Critical one-tail	1.81		1.812		1.81
P(T<=t) two-tail	0.89		0.143		0.37
t Critical two-tail	2.23		2.228		2.23

 

The results show that there is a no significant change in the colour of the biscuits made through shortening at a significance level of 5 %. The critical t value is 1.81 which is greater than the t statistic of 0.14. A similar trend is obtained for the biscuits made using coconut oil and butter. Hence there is no significant change in the colour of the biscuits from week 0 to week 2.

Table 12: Paired t-test for biscuit aroma

t-Test: Paired Two Sample for Means
	week 0	week 2
	163	987	798	314	502	756
Mean	6.663636	5.3	7.354545	6.718182	8.318182	9.036364
Variance	10.63455	10.11	10.40473	8.403636	11.29364	7.276545
Observations	11	11	11	11	11	11
Pearson Correlation	0.382295		0.704742		0.410058
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	1.263309		0.889618		-0.71379
P(T<=t) one-tail	0.117568		0.197284		0.245839
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.235137		0.394569		0.491678
t Critical two-tail	2.228139		2.228139		2.228139

The probability values are greater than the significance level which indicates that there is no significance change in the biscuit aroma in the period of two weeks.

Table 13: Paired t-test for rancidity aroma

t-Test: Paired Two Sample for Means
	Week 0	Week 2
	163	987	798	314	502	756
Mean	1.609091	4.8	1.172727	5.181818	1.727273	3.940909
Variance	2.786909	11.36	1.304182	16.83164	2.646182	10.90641
Observations	11	11	11	11	11	11
Pearson Correlation	0.377844		0.624397		0.375317
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	-3.36438		-3.79364		-2.37949
P(T<=t) one-tail	0.003595		0.001761		0.019323
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.007189		0.003522		0.038646
t Critical two-tail	2.228139		2.228139		2.228139

 
In this case, the p value is less than the significance level. Therefore, we conclude that there is a significant change in the rancidity aroma of the biscuits made using all the three type of fats.

Table 14: Paired t-test for biscuit hardnessr

t-Test: Paired Two Sample for Means
	163	987	798	314	502	756
Mean	9.354545	7.263636	7.981818	6.645455	10.35455	9.227273
Variance	2.592727	5.636545	5.649636	7.398727	2.420727	6.816182
Observations	11	11	11	11	11	11
Pearson Correlation	-0.11505		-0.3785		0.015845
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	2.297726		1.046355		1.238824
P(T<=t) one-tail	0.022213		0.160015		0.12185
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.044426		0.32003		0.243701
t Critical two-tail	2.228139		2.228139		2.228139

 

The results indicate that there is a significant change in the hardness of the biscuits made through shortening since the p value is less than the significance level. However, the biscuits made using butter and coconut oil show no significant change in hardness as indicated by the p value that is greater than the p value.

Table 15: Paired t-test for biscuit crispness

t-Test: Paired Two Sample for Means
	163	987	798	314	502	756
Mean	9.072727	8.627273	9.027273	8.472727	10.33636	10.41818
Variance	5.460182	5.042182	7.872182	4.590182	6.888545	5.589636
Observations	11	11	11	11	11	11
Pearson Correlation	0.074103		0.660402		0.690111
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	0.473763		0.864835		-0.13717
P(T<=t) one-tail	0.322921		0.203698		0.44681
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.645843		0.407397		0.89362
t Critical two-tail	2.228139		2.228139		2.228139

The results indicate that there is no significant change in the crispness of the biscuits for all the three cases from week 0 to week 2.

Table 16: Paired t-test for biscuit sweetness

t-Test: Paired Two Sample for Means
	163	987	798	314	502	756
Mean	7.736364	6.4	8.209091	8.25	7.481818	7.327273
Variance	7.010545	6.438	5.604909	8.5375	7.375636	8.418182
Observations	11	11	11	11	11	11
Pearson Correlation	0.309757		0.435128		0.506815
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	1.454434		-0.04761		0.18345
P(T<=t) one-tail	0.088242		0.481483		0.429056
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.176483		0.962967		0.858111
t Critical two-tail	2.228139		2.228139		2.228139

 

The p values for the three cases are greater than the significance level. This indicates that there is no significant change in the sweetness of the biscuits made from the three different fat types.  

Table 17: Paired t-test for vanilla flavour

t-Test: Paired Two Sample for Means
	163	987	798	314	502	756
Mean	5.190909	5.745455	6.218182	6.509091	5.945455	6.072727
Variance	11.64291	8.764727	12.05764	14.36091	9.966727	10.43818
Observations	11	11	11	11	11	11
Pearson Correlation	0.426502		0.4637		0.510539
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	-0.53563		-0.25591		-0.13355
P(T<=t) one-tail	0.301962		0.401605		0.448204
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.603923		0.80321		0.896408
t Critical two-tail	2.228139		2.228139		2.228139

 

The p value is greater than the significance level for all the three cases. This indicates that there is no significant change in the vanilla flavour of the biscuits over the period of two weeks.

Acceptability

Table 18: Paired t-test for acceptability

t-Test: Paired Two Sample for Means
	163	987	798	314	502	756
Mean	10.71818	9.5	10.64545	8.672727	9.890909	10.24545
Variance	3.575636	3.082	5.120727	2.792182	5.358909	6.554727
Observations	11	11	11	11	11	11
Pearson Correlation	0.730799		0.04955		0.634321
Hypothesized Mean Difference	0		0		0
df	10		10		10
t Stat	3.006722		2.383033		-0.56092
P(T<=t) one-tail	0.006596		0.019206		0.293599
t Critical one-tail	1.812461		1.812461		1.812461
P(T<=t) two-tail	0.013191		0.038413		0.587199
t Critical two-tail	2.228139		2.228139		2.228139

 

The p value for the biscuits made from butter and shortening is less than the significance level. This indicates that there is a decline in the acceptability of the biscuits from week 0 to week 2. However, the p value for the biscuits made using coconut oil is greater than the significance level. This shows that these biscuits have not changed significantly and can still be accepted.

The results clearly show the effect of different fat types on the sensory properties of biscuits. From the Anova test, we observe that properties of biscuits such as colour, crispness, sweetness have almost the same average rating which indicates the fat/oil type used in the baking of the biscuits has no significant effect on these properties. In addition, there is no significant change in these properties over the period of 2 weeks. On the other hand, properties such as the rancidity aroma and hardness of the biscuits change considerably among the three types of cookies. Besides, these properties also change over the two week period with the average rating declining in week 2.

The results show that the different types of fat used in this laboratory exercise do not have a significant effect on most of the sensory properties of biscuits. The only sensory properties affected significantly by the fat types are hardness and rancidity aroma. Most of the biscuit properties do not change in the two week period. The shelf life of the produced biscuit is considerably long as evidenced by the little change in most of the sensory properties.

 

Adeola, A. A., & Ohizua, E. R. (2018). Physical, chemical, and sensory properties of biscuits prepared from flour blends of unripe cooking banana, pigeon pea, and sweet potato. Food Science & Nutrition, 6(3), 532-540. doi:10.1002/fsn3.590

Caruso, M. C., Galgano, F., Colangelo, M. A., Condelli, N., Scarpa, T., Tolve, R., & Favati, F. (2016). Evaluation of the oxidative stability of bakery products by OXITEST method and sensory analysis. European Food Research and Technology, 243(7), 1183-1191. doi:10.1007/s00217-016-2831-9

CHOPRA, N., DHILLON, B., RANI, R., & SINGH, A. (2018). Physico-Nutritional and Sensory Properties of Cookies Formulated with Quinoa, Sweet Potato and Wheat Flour Blends. Current Research in Nutrition and Food Science Journal, 6(3), 798-806. doi:10.12944/crnfsj.6.3.22

Civille, G. V., & Oftedal, K. N. (2012). Sensory evaluation techniques — Make “good for you” taste “good”. Physiology & Behavior, 107(4), 598-605. doi:10.1016/j.physbeh.2012.04.015

Devi, A., & Khatkar, B. S. (2016). Physicochemical, rheological and functional properties of fats and oils in relation to cookie quality: a review. Journal of Food Science and Technology, 53(10), 3633-3641. doi:10.1007/s13197-016-2355-0

Devi, A., & Khatkar, B. S. (2017). Effects of fatty acids composition and microstructure properties of fats and oils on textural properties of dough and cookie quality. Journal of Food Science and Technology, 55(1), 321-330. doi:10.1007/s13197-017-2942-8

Gebreselassie, E., & Clifford, H. (2016). Oxidative Stability and Shelf Life of Crackers, Cookies, and Biscuits. Oxidative Stability and Shelf Life of Foods Containing Oils and Fats, 461-478. doi:10.1016/b978-1-63067-056-6.00012-4

Joseph, P. (2016). Oxidative Stability and Shelf Life of Bulk Animal Fats and Poultry Fats. Oxidative Stability and Shelf Life of Foods Containing Oils and Fats, 233-249. doi:10.1016/b978-1-63067-056-6.00006-9

Mamat, H., & Hill, S. E. (2012). Effect of fat types on the structural and textural properties of dough and semi-sweet biscuit. Journal of Food Science and Technology, 51(9), 1998-2005. doi:10.1007/s13197-012-0708-x

Mieszkowska, A., & Marzec, A. (2015). Structure Analysis of Short-Dough Biscuits and Its Correlation with Sensory Discriminants. Journal of Texture Studies, 46(5), 313-320. doi:10.1111/jtxs.12130

Obaroakpo, J., Iwanegbe, I., & Ojokoh, A. (2017). The Functional and Sensory Evaluation of Biscuits Produced from Wheat, Defatted Soybean and Coconut Flour. Current Journal of Applied Science and Technology, 23(6), 1-7. doi:10.9734/cjast/2017/30836

Rigo, M., Tozatti, P., Bezerra, J. R., Farinazzo, F., Santa, O. R., & Córdova, K. R. (2014). Chemical and sensory evaluation of formulations of biscuits with okara. Ambiência, 10(1). doi:10.5935/ambiencia.2014.01.12

Thomas, P. E. (2018). Physio-chemical and Sensory Properties of Biscuits Fortified with Carrot Powder. International Journal of Food and Fermentation Technology, 8(2). doi:10.30954/2277-9396.02.2018.12

Wirkowska-Wojdy?a, M., Bry?, J., Górska, A., & Ostrowska-Lig?za, E. (2016). Effect of enzymatic interesterification on physiochemical and thermal properties of fat used in cookies. LWT, 74, 99-105. doi:10.1016/j.lwt.2016.07.040