The relationship between prisoners’ educational motives and previous incarceration, sentence length, and sentence served

The aim of this study was to examine Norwegian prisoners’ educational motives, and how previous incarceration, sentence length, and sentence served influence such motives. Three motive categories emerged: future planning (Factor 1), social reasons and escapism (Factor 2), and competence building (Factor 3). Among prisoners who participated in education, longer sentence predicted high score on both Factor 1 and Factor 3. Also, with the greater proportion of sentence served, participants scored higher on Factor 3. Among non-participants, longer sentence predicted high score on Factor 3. Previous incarceration was not significantly related to the educational motives for either of the two groups.


Introduction
Training and education for prisoners constitutes an important but often neglected aspect of adult learning. At the European level, the European Prison Rules (Council of Europe, 2006) state that it is a duty for prison services, politicians, and policy makers to guarantee that prisoners' rights to education are met. In Norway, access to education is a fundamental civil right, also for prisoners. This is determined by Norwegian law as well as international conventions and recommendations. The prisons and prison educators are legally required to promote and support prisoners' participation in education. In order to establish a sound knowledge base that can inform both policy and practice, and help give priority to the allocation of resources for prisoners' education and training, the current study was carried out to determine the educational motives of Norwegian prisoners. This article outlines the importance that prisoners give to various motives for starting an educational programme in prison and highlights whether previous incarceration, sentence length, and sentence served have influence on the motives.

Prison education benefits
Several studies show that education has been attributed as an important factor for rehabilitation and in aiding reduction of recidivism among prisoners (Steurer and Smith, 2003;Cecil et al., 2000). A recent meta-analysis provided additional support to the premise that prison education reduces the prisoners' risk of re-offending after release. On average, prisoners who participate in education programmes have a reduction in the risk of re-offending of 13 per cent compared to those who do not participate (Davies et al., 2013). The authors have estimated that for an educational programme in prison to be cost-effective, the programme would need to reduce the three-year incarceration rate by between 1.9 and 2.6 percentage points to break even.
Thus, it is important to encourage prisoners to start an education. Substantial knowledge of the relationship between their educational motives and aspects of the sentence will help prison educators to express credible encouragements.

Prisoners' motivation for education in prison
Motivation refers to the process whereby goal-directed behaviour is instigated and sustained (Schunk et al., 2010). Within education, motivation is considered crucial to learning and achievement. Further, motivation is a complex process that can be affected by personal factors (e.g. individuals' thoughts, beliefs, and emotions) and contextual factors, such as classrooms, peer groups, and community and home influences (Schunk et al., 2010). Prisons form a very specific learning environment, with challenges differing from those faced in the ordinary school system. Thus, the prisoners' educational motives must be considered in relation to both specific personal and contextual factors, such as learning difficulties, level of education, and various aspects of their sentences (Skaalvik et al., 2003).
In previous research, prisoners taking part in education have expressed considerable motivation for study . However, in the context of being incarcerated, some of the prisoners' motives differ from those of the general population (Forster, 1981). Often these motives derive from a need to free oneself from a monotonous prison life, or as a matter of re-entry preparation prior to release (Costelloe, 2003;Parsons and Langenbach, 1993). As such, education may serve other intentions unique to the prisoner's situation than solely being attractive in itself. Also  and Skaalvik et al. (2003) identified one broad category of prisoners who attended education in order to avoid aspects of prison life, but also categories consisting of prisoners who were concerned with the value of education and post-release success or learning for its own sake. Likewise, a study by Stana et al. (1993) found that prisoners were most frequently motivated to participate in education for reasons of self-improvement and enhancement of post-release success. The least stated motive was to compensate for boredom or filling time. In line with such findings, a Canadian study demonstrated that among North American adults the most frequently cited motive for studying was that of vocational benefits (Tuijnman and Boudard, 2001).

Variables with potential influence on prisoners' motives
Prisoners vary in terms of characteristics such as previous incarceration, sentence length, and sentence served, along with gender, age, level of education, and learning difficulties. These are variables that may influence the prisoners' educational motives. In reviewing the prisonresearch literature (using ERIC, psychINFO, and Google Scholar), no other studies were found that directly aim at examining how all these variables combined relate to educational motives among prisoners. However, a population study among all Norwegian prisoners found that those serving sentences over five years were more motivated for prison education, due to both future concerns and the opportunity to build their competence, compared to those who were incarcerated for less than one year . Other studies have shown that time spent in prison is positively related to the hours of participation in prison-education programmes, as well as to the number of programmes participated in (Dhami et al., 2007;Adams et al., 1994;Glover and Lotze, 1989). These findings were partly explained by the fact that long incarceration time increases prisoners' opportunities to participate, but also as a result of adaptation and adjustment to imprisonment, as well as aging and/or prison maturation. In line with Costelloe (2003), such findings may indicate that, with time, prisoners may experience some sort of motivational change or transformation of motives. Yet another study found that more first-time prisoners participated in education than did recurrent prisoners (Souza and Dhami, 2010). There were no significant differences between the two groups in terms of sentence length and time served for their current sentences.

Prison education in Norway
The Norwegian prison system practises an import model (e.g. Christie, 1970) for the distribution of services to prisoners. Accordingly, it is the general education system that is responsible for the provision of education in prison. Given the universal interpretation of the Educational Act, prisoners have access to education in the same way as other citizens and residents. This involves seven years of mandatory primary school (age 6−13), three years of mandatory lower-secondary school (age 13−16), and three years of upper-secondary school (age 16−19). The latter is not mandatory, but is a right to which pupils can apply for general or vocational studies. Adults also have the right to a 'second chance' or supplementary basic education, and/or special education. Today, all Norwegian prisons offer education at mandatory level and at upper-secondary level. Prisoners also have access to higher education, which is anything beyond upper-secondary school. However, prison teachers do not have the required competence at this level, and can only offer unqualified help and support.
As reported by the Norwegian Ministry of Justice and Security (personal communication, 19 December 2012), there are approximately 3,700 prisoners in Norwegian prisons at any given time. Recent figures show that while 54 per cent of the prisoners participate in education, more than 80 per cent wish to start an education . The most common reasons for not participating are short sentence length, a preference for work in prison, the education of interest is not offered, or they have not been given enough information about their educational options . The two studies indicate that prisoners who do not participate in education also have educational motives, but variables such as sentence length, age, and educational challenges may be obstacles to participation. Despite an increase in prison-education participants over the last decades Diseth et al., 2006), the participation rate is still considered too low. This causes particular concern, especially as a great number of prisoners already suffer from low educational attainment.

Research problems
Our two main aims were, firstly, to replicate the categories of educational motives among Norwegian prisoners reported by , and secondly, extending this work to examine how the educational motives relate to various aspects of the prisoners' sentences. This includes previous incarceration, sentence length, and how much of the sentence had been served. Further, prisoners' gender, age, level of education, and perceived learning difficulties in reading, writing, and arithmetic may also have an impact on their motives to participate in education in prison. As such, these variables will be controlled for in the study. Three motive categories were predicted, in line with the findings of previous research reported in 2010 and 2013.

Participants
The participants were part of a target group that included all prisoners with Norwegian citizenship over 18 years of age in Norwegian prisons (individuals below 18 are not supposed to be in prison in Norway). The study was carried out in one week in October 2012. Data was collected through a standardized questionnaire. As reported by the Norwegian Ministry of Justice and Public Security (personal communication, 24 October 2012), there were a total of 2,439 prisoners with Norwegian citizenship in prison during the time period in question. Of those who received the questionnaire 1,276 completed and returned it. This represents a 52.3 per cent response rate. Of the participants 94.2 per cent were men. The mean age of the prisoners was 36 years (SD=11.6). Seven per cent had left school before compulsory school leaving age, and a total of 52.1 per cent had compulsory school as their highest level of completed education. Further, about half of the prisoners served a term of imprisonment of up to 12 months. Previous incarceration accounted for 64 per cent of the prisoners. Furthermore, 40.5 per cent of the prisoners had served less than a third of their sentence.
Due to language obstacles, prisoners of approximately one hundred other nationalities were not included, but a separate study was later carried out among three of the largest groups of foreign prisoners in Norwegian prisons (Eikeland et al, 2014).

The questionnaire
The questionnaire was designed partly to assess the prisoners' educational background, educational motives, and actual participation in prison education, as well as different features of their sentences and their perceptions of difficulties in reading, writing, and arithmetic. Standard demographic data was also requested, such as gender and age.
Central to the present study was an educational-motives scale, originally developed by Skaalvik et al. (2003), which has been applied in previous national surveys among prisoners in Norway . With the permission of the original authors, the scale has been modified over time. The present version consisted of 15 potentially viable reasons for educational participation in prison. The prisoners, both education participants and non-participants, were requested to specify how important each of these reasons was or would be for their choosing to participate in education. Examples of items include 'Because it is better than working in prison', 'To learn about a subject', and 'To be better able to cope with life after release'. Each item was accompanied by four response categories ('very important', 'important', 'less important', and 'not important'), and the prisoners were asked to mark the appropriate ones. Data was analysed utilizing Cronbach's α to examine the internal consistency of the educational-motives scale. The analysis generated a reliability coefficient of 0.87.
The part of the questionnaire that included aspects of the prisoners' sentences − a section that was of particular interest for the research − covered previous incarceration, sentence length, and sentence served. Sentence length included 15 options, including 'three months or less', 'three to six months', 'six to twelve months', 'one to two years', 'two to three years', and onwards up to 'more than twelve years'.
Sentence served was grouped into: 'just started', 'shorter than a third', 'between a third and two-thirds', 'over two-thirds', and 'practically done'. Previous incarceration was provided with a 'yes' or 'no' option.
Further, level of education was divided into eight categories: • 'not completed any education' • 'primary school/lower-secondary school' • 'one year of upper-secondary school' • 'two years of upper-secondary school' • 'completed upper-secondary school' • 'vocational college' • 'individual subjects at a university or university college' • 'a degree course at a university or university college'.
With regard to self-reported reading, writing, and arithmetic difficulties, the prisoners had four perceptual categories from which to select: 'yes, to a great extent', 'yes, to some extent', 'yes, but just a little', and 'no, not at all'. The reliability and validity of the questionnaire had also been tested in advance in a pilot among prisoners in one prison in Norway.

Procedure and ethics
Serving the Ministry of Education, the County Governor of Hordaland, Department of Education, is delegated the overall national responsibility for prison education. In order to accomplish the study, one official from this department contacted each prison governor and each headmaster in charge of prison education and provided them with a briefing on the study's objectives and guidance on assessment. In addition, an explanatory letter of the procedures was forwarded to the same persons. In compliance with instructions set forth by the research group, the prison governor or the headmaster in charge of education managed the survey data collection. The study was approved by the Ombudsman for Privacy in Research, the Norwegian Social Science Data Services. Additional consent was granted by the prison authorities and the Ministry of Justice and Public Security. Along with the questionnaire, the prisoners were provided with written information that clarified the purpose and procedure of the study. They were assured of their anonymity and the confidentiality of data. It was emphasized that participation was voluntary, with the right to withdraw at any time during completion of the form, with no subsequent consequences. In case of literacy difficulties, the prisoners were offered assistance by prison personnel in completing the questionnaire.

Results
The educational-motives scale was factor analysed for the 1,276 respondents. We decided to use an exploratory approach to data reduction as the present study, in contrast to previous studies, targeted only prisoners holding Norwegian citizenship. For this purpose, a principalaxis factoring analysis for the extraction of factors with oblique rotation was used. This datareduction technique was chosen over the more traditional principal-component analysis with varimax rotation as it is considered to be more suitable for producing a factor structure (Russell, 2002). The procedure allows for correlated factors, which often give a more accurate and realistic representation of the data. However, a principal-component analysis with varimax rotation was also performed as potential support to any findings.
The Eigenvalue above-one criterion produced a three-factor solution, which accounted for 58.2 per cent of the variance. According to the guidelines for identifying significant factor loadings based on sample size (Hair et al., 1995), seven items, five items, and three items, respectively, were considered in the interpretation of the first, second, and third factor ( Table 1). The factors were labelled 'future planning' (Factor 1), 'social reasons and escapism' (Factor 2), and 'competence building' (Factor 3). The factors had initial eigenvalues of 5.4 (Factor 1), 2.0 (Factor 2), and 1.3 (Factor 3). They explained 36.2 per cent, 13.5 per cent, and 8.5 per cent of the variance, respectively. Hence, Factor 1 emerges as the most clear-cut compared to the other factors. The inter-factor correlation between the three factors ranged from 0.22 to 0.47. Additional support for a three-factor solution was obtained by reviewing the scree plot, which displayed a notable change of slope after the third factor. Also, the principal component analysis yielded the same results. The three factors are presented in Table 1. In addition, the mean values on each motivational item among the prisoners are shown in Figure 1. Note: Structure matrix. Oblique rotated factor loadings from the principal axis factor analysis (n=1,276). Primary factor loadings italicized; h 2 =extraction communalities (estimates of variances in each item accounted for by the factors). The three factors were transformed into three sum indices. The correlations between the indices were 0.41 (Factors 1 and 2), 0.52 (Factors 1 and 3), and 0.32 (Factors 2 and 3). Cronbach's α coefficients were 0.86, 0.77, and 0.71 for the future planning index, the social and escapism index, and the competence building index, respectively. In order to examine previous incarceration, sentence length, and sentence served as predictors of the prisoners' educational motives, three simultaneous regression analyses were performed. The motive indices were entered as dependent variables in the separate analyses. For better interpretation of the results, the indices had been reversed in advance. Previous incarceration, sentence length, and sentence served were entered as independent variables in each analysis. Sentence length was coded with 15 values (see 'The questionnaire', above). Sentence served was assigned with three values: 1=shorter than a third; 2=between a third and two-thirds; and 3=over two-thirds, and previous incarceration was entered as 1=yes and 2=no.
The following variables were entered as control variables: gender (male=1; female=2); age (a continuous variable); level of education (1=no education; 2=primary/lower-secondary school, and one or two years of upper-secondary school; 3=completed upper-secondary school or vocational education; 4=individual subjects or a degree course at a university or university college); and perceived difficulties in reading, writing, and arithmetic (1=no, not at all; 2=yes, but just a little; 3=yes, to some extent; and 4=yes, to a great extent). All analyses included both prisoners who participated in education and those who did not. Table 2 shows the effect of each independent variable on the prisoners' educational motives, with all other variables kept constant. For the sample as a whole, three independent variables emerged as being significantly related to the motive categories: sentence length, age, and reading difficulties (Table 2). Prisoners with longer sentences are more likely than those serving shorter sentences to see education as important in terms of both Factor 1 and Factor 3. Also, with decreasing age, these motives are considered more important. Finally, prisoners with a greater extent of reading difficulties are more likely to consider education as important for reasons linked to Factor 2 than those with less difficulty or no such problems.
In order to examine if the independent variables predicted the motive categories differently for prisoners who participated in education and those who did not the three regression analyses were also conducted separately for these two groups.
For participants, three independent variables were significantly related to the motive categories: sentence length, sentence served, and age. With regards to sentence length, the results are consistent with that of the total sample. Further, the greater the proportion of sentence served, the more likely prisoners in education are to emphasize the importance of education from a future planning perspective (B=0.10, β=0.10, p=<0.05). Lastly, younger participants score like the total sample on Factor 1.
For prisoners who did not participate in education, four independent variables were significantly related to the motive categories: sentence length, age, reading difficulties, and writing difficulties. Equivalent to the results of the sample as a whole, the longer the sentence the more likely non-participants are to value education for reasons connected to Factor 3. With decreasing age non-participants are more likely to see education as important for Factors 1 and 3 (like the total sample) but also for Factor 2 (B=-0.01, β=-0.12, p=<0.05). Moreover, non-participants with a great extent of reading difficulties emphasize Factors 1 and 3 to a higher degree than nonparticipants with less or no such difficulties (B=0.20, β=0.18, p<0.05; B=0.15, β=0.17, p=<0.05, respectively). However, non-participants with a great extent of writing difficulties are less likely than other non-participants to value Factor 3 (B=-0.14, β=-0.18, p=<0.05).

Discussion
Prisoners experience many important changes in their lives and circumstances that influence the development of their motives for engaging in education. In the present study of Norwegian prisoners, both those who participated in education and those who did not were requested to consider the importance of 15 reasons for their choosing to participate in education while incarcerated. Three motive categories surfaced from the factor analysis: 'future planning' (Factor 1), 'social reasons and escapism' (Factor 2), and 'competence building' (Factor 3). Of these, Factor 1 was by far the most important educational motive. Further, several cross loadings did appear between factors at the item level, particularly concerning Factors 1 and 3. The correspondence between future planning and competence building is not so unexpected, as prisoners may consider competence building as an integrated element of preparing for re-entry to the community, and re-entry to the community as an integrated element of competence building.
The emergence of three motive categories is coherent with findings in previous Norwegian studies examining samples of prisoners from a number of nations . Likewise, in a Swedish study Eriksson and Samuelson (2007) factor analysed prisoners' motives for educational participation. The results were closely consistent with our findings, with the exception of identifying a fourth category, 'social and emotional needs', although in the present study, the motive of social reasons and escapism overlaps with this motive. Such a motive is also related to the motive that was most frequently cited by prison students in Costelloe's (2003) Irish study of third-level education, namely to alleviate boredom, and Forster's (1990) view of prisoners taking up education to break free from the constraints of prison life. According to Forster, motives like this apply only rarely outside a prison context, and from a learning point of view, they may be regarded as 'negative' when education is not seen as attractive in itself. However, many prisoners' negative experiences of education in childhood have to be borne in mind. According to the Council of Europe (1990), motivation must therefore be seen as a dynamic concept, and what appears as low motivation among prisoners can be understood as a result of past experiences. In line with Costelloe (2003), 'escape motives' may also lead to transformation of educational motives into more adequate ones that can promote learning.

Prison sentence variables influencing educational motives
After controlling for other factors related to the educational motives, it was found that the prisoners who served long sentences were more likely to value education for reasons that can be associated with future planning and competence building. When analyses were conducted separately for prisoners who participated in education and non-participants, both motive categories turned out as important for participants but only competence building for nonparticipants. The results indicate that participants may have had clear intentions to change their future when they started education or that, during education, they have realized its importance for coping better with life after release and avoid re-offending. However, it must be noted that from a foreign perspective, long sentences in Norway are normally fairly short, as five years and beyond are considered long-term imprisonment, with a maximum sentence of 21 years (in severe criminal cases offenders may, however, be convicted to 'special detention', which may in some very few cases correspond to that of life-long incarceration in other correctional systems). Hence, Norwegian prisoners are bound to position themselves toward a post-release scenario at a relatively early stage of their incarceration. Accordingly, this may influence their educational motives, such as not giving much value to filling time and instead emphasizing the importance of competence building and preparation for social reality in the not too distant future. In contrast to prisoners who participate in education, non-participants with long sentences who value competence building may have met barriers to starting an education in prison , or they are considering this option.
The more of their sentence the participants had served, the more likely they were to emphasize the importance of education from a future planning perspective. In a US study, Jackson and Innes (2000) also found that prisoners who had served at least half of their sentence were more likely to participate in educational programmes. A possible interpretation of this is that, as they approach release, the prisoners are inclined to experience a reality check regarding the importance of education for a successful re-integration to society. The association between the motive of future planning and time served was not true for the non-participants. Although contrary to previous findings (Souza and Dhami, 2010), it would be a reasonable assumption that, with recurring imprisonment, prisoners would be more susceptible to consider education as a necessity for post-release success and thus emphasize the motive of future planning or competence building. However, previous incarceration had no influence on any of the motive categories. Finally, sentence length, sentence served, and previous incarceration did not affect the social reasons and escapism motive for either participants or non-participants.

The impact of other variables
Of the control variables, both age and difficulties in reading, writing, and arithmetic had a significant influence on the motive categories. Younger participants were more likely than older participants to value future planning, while younger non-participants scored higher than older non-participants on all three categories.  found that younger prisoners were more likely than older prisoners to emphasize education for reasons linked to preparation for life upon release. As low educational attainment is especially prevalent among younger prisoners , it is likely that these prisoners may reflect more than older prisoners on their future prospects and the alternatives they face, for which education becomes an important influence. Also, when they do not participate in education, younger prisoners may have stronger social needs or a need to escape from their present situation. Generally, with increasing reading difficulties, prisoners are more likely to consider social reasons and escapism as a motive for starting an education. There is also a tendency that the more reading problems non-participants have, the more they value future planning or competence building compared to non-participants with less such difficulty. However, for non-participants with writing problems, the opposite is true for competence building, which makes it difficult to explain such findings.

Practical implications, limitations, and future research
The study shows that prison educators need to support prisoners who position themselves towards a crime-free post-release scenario. A possible consequence of the fact that prisoners with short sentences are less likely to see education as important for future planning and competence building is that short sentences become an impediment to participation in education. Because many young prisoners serve short sentences, low motivation for education and work can easily become a stepping stone to further criminality. Thus, it is important to enable their completion of short courses that can increase their competence upon release. Novitzky et al. (2013) evaluated an intensive English and maths delivery model of 45 hours over a period of three to four weeks and suggest that working intensively is particularly useful for prisoners who are serving short sentences. One should also be aware of the many young prisoners who do not participate in education but do consider education as important for reasons affiliated with important motive categories. Such circumstances create a fertile ground for the prison educators to help and guide them into education.
The study has several limitations. The results may have been influenced towards a more 'official' direction because the survey was administered by prison and prison-education staff. One might also argue that different results would have been obtained when using additional control variables, such as the subjects' socio-economic background, type of crime, drug problems, and information about mental disorders. However, due to ethical considerations it was not possible for us to collect these data, but future research may have the opportunity to control for such variables. We should also acknowledge that the educational-motives scale does not reflect all common reasons to take part in education. There is consequently a need for more research, qualitative and quantitative, that further explores the dynamic concept of educational motivation in prison. Terje Manger, PhD, is a professor in educational psychology at the University of Bergen. He has published on topics such as gender differences in mathematics, educational motives and self-concept, prison education, behaviour problems, and general educational psychology.