Introduction
Quality education is key in the development of any nation. It provides opportunities for people to develop their competencies that can allow them to perform well in society and it enhances the social, intellectual and economic growth of communities (Openheimer, 2010).
Education must bring the possibility to face continuous changes and professional competitiveness. Learning is a multifactorial phenomenon that requires attention as it is dependent on many social, political and psychological elements. Many studies related to factors attributed to students´ characteristics and the effect of these on their learning process, some of the factors more mentioned are: prior knowledge (Alexander & Judy, 1988; Ausubel, Novak, & Hanesian, 2006; Dochy, 1991a; Dochy & Alexander, 1995; Hailikari, 2009; Hattie, 2009; Marzano, 2004; Meltzer 2002; Roschelle, 1995; Sagastizabal, Perlo, Pivetta & San Martín , 2009; Shapiro, 2004; Thompson & Zamboanga, 2004), social factors (Alexander & Judy, 1988; Dochy, 1991a; Tinto, 1992), cultural factors (Sagastizabal et al., 2009), individual attributes like gender and race (Hattie, 2009; Tinto, 1992), interests and intentions from student (Hattie, 2009), commitment (Hattie, 2009; Tinto, 1992), motivation (Alexander & Judy, 1988; Boiché, Sarrazin, Grouzet, Pelletier & Chanal, 2008;Garris, Ahlers, & Driskell, 2002; Graham & Weiner, 1996; Hattie,2009; Wong, 2012), economic factors (Sagastizabal et al., 2009; Tinto, 1992), self-efficacy (Bandura, 1971, 1982, 2006; McKenzie & Schweitzer, 2001; Zimmerman, 1989), stress and anxiety (Hattie, 2009), communication and social skills (Porter, 2008), personality (Porter, 2008), emotions (Kleres, 2010),time spent to study (Grouws & Cebulla, 2000), learning approaches from student (López, Esteban & Pérez, 2006), metacognition and prior academic development (Dochy, 1991b), and more of them that could be listed.
The Coleman report (1966) analyses education in public institutions in the USA through a qualitative study, including types of teachers, students, principals, and he says that the most important factor influencing academic achievement is what students bring to school with them such as prior education, family background, culture and interests (Coleman, Campell Hobson, McPartland, Mood, Weinfeld & York, 1966 ). They conclude that the majority of the contribution in the variance for academic development in students comes from those personal features.
Thus learning should be significant and should be studied as a process that includes the interaction of the new knowledge to be learned and the knowledge subjects already posses. This research takes learning as a process of active construction of knowledge (Ausubel, Novak, & Hanesian, 2006). Among the problems observed in education we have low grades, high failure rates and high dropout indices, large groups when students´have inconsistencies in their prior knowledge (Fernández, Mena & Riviere,2010;González, Castañeda Maytorena, 2000; Palacios Andrade, 2007;Stinebrickner & Stinebrickner, 2013;Tinto, 1992). Ibáñez (2007) for example, suggests that the main causes of low school achievement are: missing study habits, low motivation from students, bad teaching practices, incorrect politics and unfit education models.
In this chapter, the authors take the perspective of learning as a process, the relation to change from knowledge existing in the memory interacting with the new knowledge to be acquired. Information processing theory considers prior knowledge as an assimilation environment and states that new material is related and integrated to it (Dochy, 1991a). Ausubel (2006) when discussing assimilation theory maintains that acquiring new information depends mainly from the preexisting ideas in the cognitive structure, and that meaningful learning would be determined by those ideas. Thus it is important to study prior knowledge as a determinant of new learning (Alexander & Judy, 1988; Ausubel, Novak, & Hanesian, 2006; Dochy, 1991a; Dochy & Alexander, 1995; Hailikari, 2009; Hattie, 2009;Karabel & Halsey, 1977; Manzano, 2004; Thompson & Zamboanga, 2004).
Hailikari (2009) established that one of the first researchers who affirmed the relevance of prior knowledge on the learning process was Bloom in 1956, when he wrote that the learning process is determined mainly by the cognitive behaviors or prerequisites, term used referring to prior knowledge. Bloom also mentioned that more than a half of the variance of learning outcomes depends from prior knowledge.
In the same direction, Ausubel, Novak y Hanesian (2006) pointed that if they had to reduce education to a sentence, the main factor influencing learning, is what the learner already knows, and that professors must consider this to teach consequently. Dochy (1991a) concluded that this sentence assumes three ideas: prior knowledge is an important variable in education; content and structure of prior knowledge must be relevant to new knowledge to obtain optimum results; and the better condition of learning is obtained in base of the accordance and connection to prior knowledge.
Moreover it is important to define what we consider prior knowledge to be and what it isn’t, its different types, effects, nature and its inherent qualities in order to better understand how new knowledge gets integrated in to our cognitive structure and how it can be measured. The purpose of this project then was to develop a valid and reliable instrument that measures the specific inherent qualities selected from prior knowledge so later in an extended project we can attempt to determine the effect these characteristics have on learning.
Prior Knowledge
The main idea underlying the interaction between prior knowledge and new learning is not that prior knowledge itself leads to success or failure in learning. Instead, prior knowledge is conceived as the raw material that conditions learning. Furió and Guisasola (2001)point out that one of the most worrying aspects of prior ideas is not that they are correct or incorrect but rather the persistence of information. Larkin (1983) mentions that prior knowledge is concrete, individual, complex and it is resistance to change. Moreover Emeigh (2008) pointed out that the more complex the new concept to learn is, the greater complex cognitive structure required.
Prior knowledge is present before the implementation of any instruction or learning task, and has some proprieties such as availability, accessibility, recovery, dynamism, transferability, quantity, with structured schemes and with levels of relevance to the new learning task (Dochy & Alexander, 1995; Dochy, Moerkerke & Segers, 1999). When prior knowledge is characterized as being dynamic, it means that knowledge changes according to the passage of time and as proposed by Arroyo, Morales, Silva, Camacho, Canales and Carpio (2008) this knowledge is not controllable in a direct manner.
Furthermore, according to Hertwig and Todd (2003), the database where we obtain our inferences and predictions has limited amount and limited processing memory, also called working memory. Hernández (2012) adds that the capacity of the processing memory is limited in duration and in units of processed information called chunks or bits.
Prior Knowledge Classification
Many classifications have emerged to describe types of prior knowledge, we can establish some of the most important considered for this research. Alexander, Schallert and Hare (1991) employed one classification of subcategories in terms of the specialization of the body of knowledge, which could be considered areas or branches of information. They propose categories such as the content knowledge that refers to concepts from some general area of knowledge (e.g. scientific knowledge); discipline knowledge that is used for highly subsets of a specialized field of study (e.g. physics); and domain knowledge, mainly used to refer to subjects(e.g. electricity and magnetism or calculus). In this classification, the most general name of prior knowledge could be the content knowledge and the most specific could be the domain knowledge.
There are other classifications proposed by Dochy and Alexander (1995). They mention different states of knowledge as procedural knowledge which is related to steps, plans, rules, skills and operations, also calledas know how. Declarative knowledge, on the other hand is concernedwith concepts, symbols, meanings and definitions, also called as know that. Finallyconditional knowledge refers to the situation of the context of learning and is also known by other authors as know when and where.
Prior knowledge has a broader typology when it comes to its inherent qualities (Dochy, 1991a). Prior knowledge inherent qualities are characteristics of the different information stored in the memory. An important classification of these inherent qualities is done by Ambrose, Bridges, DiPietro, Lovett & Norman(2010), which includes: content, quantity and structure. The characteristics of content are related to the comparison of the information from the student with the knowledge defined by the science, two inherent qualities are possible: correctness or misconceptions. From the subset of quantity, another three inherent qualities are possible: completeness, incompleteness or absence. Finally, from the structure subset which consists of hierarchies of concept, relation and meta-relation types (Nguyen, Kaneiwa & Nguyen, 2010) and that it is possible to have another two inherent qualities: wrong structure or correct structure. From the inherent qualities already mentioned, the characteristics selected in this study are from the subsets of content and quantity (see table 1).
Table1
Prior knowledge inherent qualities selected for the study
Subset / Inherent Qualities / Condition of prior knowledgeCompleteness / Enough for the learning task
Quantity / Incompleteness / Not enough & without mismatches
Absence / No response
Content / Correctness / Matches with science
Misconception / Mismatches with science
The inherent qualities on the table 1 for content arecompleteness which represents enough information to succeed learning the new knowledge, incompleteness which is related to some information missing and to the amount of information present (some correct but not complete) and absence that represents no prior knowledge exposed by the student; for the quantity subset there can be misconceptions when the information is incongruent with what is accepted by society or science and correctness which is the characteristic where the prior knowledge matches with the information accepted by the science. In addition, as it was said in the description of knowledge, the prior knowledge has other inherent qualities that are not considered in this study such as availability, accessibility, durability and relevance.
Another classification of prior knowledge is depending on its nature that can be observable or not (Dochy & Alexander, 1995; Reber, 1989). It can be explicit for knowledge that can be observed or measured (e.g. when a student state the response of a question in a test), and can be tacit when knowledge cannot be clearly observed or identified (e.g. the process that a student used to get to the explicit answer that is not expressed in).
Finally one important classification done by Ambrose, Bridges, DiPietro, Lovett, and Norman, (2010) is related to the effects that prior knowledge could have with the new information to be learned. The effects mentioned by them are the ones that facilitate and contribute in the learning process commonly referring to completeness and correctness,while the others are the ones that interferein the acquisition of new information like misconceptions, and it is not very clear which is the effect of absence and incompleteness of prior knowledge.
There are three types of learning depending on the conditions of the inherent qualities of prior knowledge. Thefirst condition is when the student doesn´t have any prior knowledge related to the specific topic to be learned. The process of learning involvedwith the absence of prior knowledge consists in the acquisition of new knowledge. The second condition is when students have some knowledge contents that are correct, but not complete. This kind of learning involves a gap filling process (Carey, 1999). The third condition is when prior knowledge has conflicting information with the new knowledge to be learned, or when prior knowledge has misconceptions. This process of learning requires a conceptual change (Chi, 2008).
Prior Knowledge Assessment
Now that we presented some prior knowledge categorizations and some of the learning process involved, we can turn to the design of a test to measure students´ prior knowledge. Due to the many types of knowledge described, it is necessary to select the types that are relevant or at least measurable. An explanation and justification of the inherent qualities already selected is required to clarify the study. First, a distinction of procedural knowledge and declarative knowledge has to be separated in different tests to identify the distinction of processes and concepts. These was identified in a prior research (Arellano, Mendoza & Villarreal, 2015) when a test was developed to detect the inherent qualities of the knowledge involved to solve problems for electrostatics specifically on Coulomb law on engineering students.The test had the responses already included but with some errors intentionally included, asking to the students´ to identify them. Issues emerged because of the design of the instrument that didn´t establish clearly the responses that had different inherent qualities of between declarative and procedural prior knowledge.Second, the test posed some problems to measure procedural knowledge because the items were sequenced as series and if any of the first items was answered incorrectly by the students, the later items couldn´t be answered. For these reasons, the procedural knowledge had to be set aside in this part of research and only declarative inherent qualities of the prior knowledge were considered for measurement. Similar to this problem from measuring procedural prior knowledge, the inherent qualities from the subset of structure define relations between prior knowledge concepts and operations in hierarchy, something difficult to do because of the design of the multiple choice test selected.
In order to assess declarative prior knowledge present in the memory of each student, we selected the inherent qualities of correctness, completeness, misconceptions, incompleteness, and the absence of knowledge (figure 1). The definitions of each inherent quality of prior knowledge selected used for the design of the instrument are:
Correctness is related to the consistence, acceptance and congruence of the student knowledge with the scientific knowledge (Ambrose, Bridges, DiPietro, Lovett & Norman, 2010). Completeness is related to a very close enough amount of information of declarative knowledge from the student compared to what is defined by science. It is possible to identify that the definition of correctness and completeness are very similar because they have to be equal or almost equal to what is scientifically accepted, but correctness is related to the content of the prior knowledge and completeness refers to the quantity of prior knowledge. There are coincidences and differences that are easy to detect in the conceptual definition of this two inherent qualities, but when it is time to measure them it is hard to do it by separate or mutually excluding. For measuring purposes on the test, to consider a prior knowledge as a correct definition from the student, the information must match in quantity and content with science definitions. An example of correct prior knowledge is when a student affirm that the unit of linear length in the International System of Units is the meter, because that is the scientific definition.
A Misconceptionis incorrect knowledge or the presence of errors in students´ knowledge, when the information is incongruent with what is accepted by the society or science (Ambrose, Bridges, DiPietro, Lovett & Norman, 2010; Smith, diSessa & Roschelle, 1993; Taylor & Kowalski, 2004). Some of the terms used to name the misconceptions are:distorded understanding, inaccurate prior knowledge (Ambrose, Bridges, DiPietro, Lovett & Norman, 2010), alternative conceptions (Hewson Hewson, 1983), naive beliefs (Caramazza, McCloskey Green, 1981), alternative beliefs (Hammer, 1996), false beliefs(Chi, 2008; Taylor & Kowalski, 2004), false conceptions (Duit, 1993), incorrect beliefs, misinformation, inaccurate belief, inconsistencies (Taylor & Kowalski, 2004), incongruous beliefs (Rebich y Gautier, 2005),alternative frameworks (Taber, 2001), naive theories (McCloskey, 1983; Pine, Messer & John, 2001), erroneous ideas, mistaken conceptions, misunderstandings(Thompson & Zamboanga, 2004),mistaken ideas, inadequate existing knowledge, faulty conceptions, flawed conceptualizations (Smith, diSessa & Roschelle, 1993) flawed mental models (Chi, 2008), misconceived qualitative explanations (Chi, Roscoe, Slotta, Roy & Chase, 2012).One very common example of misconception is when students´ think that earth is flat.
Incompleteness refers to some information correct and some information missing, in other words, the information is needed in its entirety as required for the new learning (Dochy, 1991a; Levesque, 1981). Incomplete knowledge has a deficit of information in quantity and complexity. Terms used to talk about knowledge that is not complete are: lack of information (Christen & Murphy, 1991; Taylor & Kowalski, 2004),critical gaps between new and prior knowledge, insufficient knowledge (Ambrose, Bridges, DiPietro, Lovett & Norman, 2010) andsignificant gap of knowledge (Ungar, 2000). An example of knowledge incomplete is when a student is asked for the definition of sine and cosine in a right triangle and he or she knows which of the sides are the legs and the hypotenuse, but doesn´t know the definition of sine and cosine.
Absence of knowledgeis the information that is not available as a mental representation in students (Smithson, Bartos, & Takemura, 2000).Another terms used to refer to the absence of knowledge are: scientifically uninformed, analfabetismo científico or scientific illiteracy, ignorance (Risbey O´Kane, 2011; Ungar, 2000), unknown or uncertainty (Moss & Schneider, 2000), effective ignorance (Risbey & Kandlikar, 2007). A case of the absence of knowledge is when the student affirm that doesn´t have any response.
Figure1.Inherent qualities of prior knowledge selected for this study adapted from Ambrose, Bridges, DiPietro, Lovett, Norman, (2010).
Problem Statement
Several studies have determined that prior knowledge affects the learning process (Alexander & Judy, 1988; Ausubel, Novak, Hanesian, 2006; Dochy, 1991a; Dochy y Alexander, 1995; Dochy & Segers, 2014; Hailikari, 2009; Hattie, 2009; Marzano, 2004; Meltzer 2002; Roschelle, 1995, 2004; Sagastizabal, Perlo, Pivetta & San Martín , 2009; Shapiro, 2004; Thompson & Zamboanga, 2004), however there has not been research conducted in order to establish clearly how each inherent quality of prior knowledge affects differently the learning process. The qualities of prior knowledge more mentioned and with particular interest are those selected in Figure 1. The premise of this study is that we could assess the amount and content of prior knowledge inherent qualities then the learning process could be improved or at least better understood.One of the objectives we set up to achieve was the design of a test with a high Cronbach alpha coefficient that would provide some information regarding prior knowledgeso later it could be correlated with learning outcomes.