Memorandum submitted by Michael Rice
BACKGROUND
Is there a relationship between reading attainment,
employment, and crime?
1. Since the Sheriff of London conducted
a straw poll of prisoner literacy at Newgate in 1808, it has been
a commonplace that low educational attainment initiates criminal
careers and that educational interventions can accelerate criminal
career termination. For a time, reformers believed that for every
school that was built, a prison could be demolished. Long after
this hope proved to be unjustified, basic skills tutors in prisons
have continued to believe that the value of their work is not
only instructional but also rehabilitative. If their belief has
any support, it may derive from that fact that teaching involves
both an instructional and a mentoring role and that, even if instruction
cannot achieve very much in the limited time that is available,
mentoring might be effective.
2. Many offenders, even if they are functionally
literate, have poor alphabetic skills. However, this is because
low reading attainment is associated with personal causal risk
factors such as low intellectual ability and early-onset behavioural
problems associated with the diagnoses of ADHD or hyperactivity-impulsivity-inattention.
Low reading attainment is also associated with environmental causal
risk factors such as low levels of language and literacy in the
childhood home and poor teaching in the infant and junior school.
3. The best test of the relationship between
reading attainment and offending involves longitudinal follow-up
of children from the general population with reading attainment
assessed in the early school years and later delinquency assessed
by several different measures. Tests of this kind undertaken by
the National Survey of Health and Development and the Institute
of Psychiatry's inner London borough study in the United Kingdom,
the Dunedin Multidisciplinary Health and Development Study and
the Christchurch Health and Development Study in New Zealand,
the Seattle Social Development Project in the United States, the
Concordia High Risk Project in Canada, and the Australian Temperament
Project all find that the association between reading attainment
and offending (or related constructs) is spurious.
4. Although those findings may seem counter-intuitional,
they converge with findings from cross-sectional studies of prison
populations. In their turn, the prison studies indicate that the
relationship between low reading attainment and offending is spurious
when socio-economic status and the social disadvantage for which
it stands proxy are taken into account. These findings are predicted
by the predominantly blue-collar nature of the prison population
and by the finding that while most people in the white-collar
population are functionally literate, most people in the blue-collar
population are not. (See Figure.) Thus, it appears to be a robust
finding that while low attainment is a risk indicator for offending
it is not a risk mechanism.
5. However, this finding may not bear on
the question of what brings about desistance from offending. There
are two reasons for this conjecture. First, the way out of a criminal
career is not necessarily the reverse of the way in. Second, the
question of what brings about desistance can be distinguished
from the question of what brings about social inclusion. While
the importance of basic skills interventions in equipping people
for civilian life is axiomatic in an advanced society, the importance
of basic skills interventions in preventing re-offending needs
empirical confirmation.
6. We do not have that empirical confirmation.
A major review of the "what works" research [11] has
concluded that whereas adult basic education is a promising method
for reducing recidivism, there is not enough evidence to conclude
that it is effective [2]. For the time being, reading instruction
in prisons can be justified not by its potential for reducing
reoffending but by its potential for restoring to society those
people who are excluded from full citizenship because they have
yet to attain functional literacy. In short, reading interventions
for offenders are justified not by reference to human wrongs but
by reference to human rights.
7. How does that affect prospects of employment?
The current prioritisation of basic skills interventions for offenders
is driven by a belief that raised attainments will also enhance
employment opportunities, and that stable employment will reduce
the risk of reoffending. However, as the probation inspectorate
has acknowledged in a recent thematic review [1], the relationship
between employment and offending is not completely straightforward.
Given, among other things, the inconclusive nature of the evidence
for a causal relationship between unemployment and crime, a finding
that desistance is associated with the quality of employment rather
than employment per se [12], a finding that first-time
criminal conviction actually has a positive effect on the legitimate
income of young offenders [5], and a possibility that work provides
a turning-point for older but not younger offenders [13], it may
be simplistic to expect that any improvement in basic skills will
lead to employment, and that employment in turn will make the
offender less likely to reoffend. Whether or not people turn their
backs on offending is likely to depend upon many personal attributes,
among which functional literacy may be neither necessary nor sufficient.
Further insight into the complexity of the relationship between
employment and reoffending is given by an impressive review for
the Department of Work and Pensions [4].
In brief
— Many offenders, even if they are
functionally literate, have poor alphabetic skills.
— However, the association between
reading attainment and offending is spurious.
— So that, while low attainment is
a risk indicator for offending, it is not a risk mechanism.
— The relationship between employment
and offending is neither straightforward nor yet fully understood.
— And so, for the time being, reading
interventions for offenders are better justified by reference
to human rights than by reference to human wrongs.
FOREGROUND
What are the purposes of reading skills assessment?
8. A brief initial screening assessment
may help to determine whether a prisoner can cope with the reading
demands of a particular rehabilitative programme, with respect
to two key dimensions of reading skill: accuracy and fluency.
(It needs to be asked whether the current initial assessment is
sufficiently informative in either respect.) By contrast, a supplementary
diagnostic assessment may establish the nature and extent of a
prisoner's reading skills deficits, permitting the tutor to devise
a curriculum for remedial teaching and setting a baseline from
which future progress can be measured. These are separate purposes,
and in current practice they are distinct.
9. There are two important issues to be
resolved. The first issue concerns the timing and location of
assessments. The second issue concerns their scope. In prisons,
both issues have to be determined by pragmatic as well as by theoretical
considerations. The best achievable course may simply be the least
unsatisfactory.
10. Currently, prison staff conduct the
initial assessment of all prisoners by means of a written group
test during the induction period. This initial assessment procedure
is likely to under-identify functionally literate prisoners, while
generating very little useful information about prisoners who
are less than functionally literate. It is also unsatisfactory,
in that it confounds reading and writing ability. For the purpose
of aggregate year-on-year comparisons this may not matter very
much, assuming that the sources of error remain constant. However,
data of such poor quality are unfit for comparisons with data
obtained individualised assessments of members of the general
population.
When and where should reading skills be assessed?
11. Assessments need to be conducted at
a time and under conditions that permit prisoners to do their
best. Although some people might propose that since real life
subjects people to all kinds of stressors a realistic assessment
would also subject people to stress, it would be both counter-productive
and unethical to demoralise so many prisoners who have already
been alienated by their childhood experience of education. Accordingly,
it is preferable to delay diagnostic assessment until prisoners
have adjusted to the experience of being sentenced and also, perhaps,
to the experience of being transferred.
12. It may be a counsel of perfection to
suggest in addition that assessment should take place in a quiet
room where no other people are present, since current staffing
levels are unlikely to make such an arrangement feasible. Nevertheless,
something of the kind is necessary if reliable baseline data are
to be obtained at the start of an intervention programme. If reliable
data are not obtained, then it is likely that regression to the
mean on re-test will exaggerate the effectiveness of the intervention;
it might even indicate a positive effect when the effect has actually
been nil or negative.
What should be the scope of reading skills assessment?
13. With respect to scope, there are two
sets of considerations. The first consideration is the learning
agenda, which is identified by exploratory assessment of the learner's
current knowledge of the alphabetic writing system and the way
in which it is used to transcribe language. This assessment needs
to take place at the start of the intervention.
14. The second consideration concerns intellectual
and motivational impediments to learning. If the learner appears
to make satisfactory progress after a few weeks, no further exploratory
assessments are needed. If progress appears to be unusually slow,
then further exploratory assessments may be indicated. In the
prison population, especially, there is a wide repertoire of intellectual
and motivational impediments to learning. It encompasses child
abuse and neglect, linguistic impoverishment in the childhood
home, low verbal ability, uncorrected visual and hearing impairments
in childhood, unskilled teaching in the junior school and mistaken
conjectures about literacy practice, closed-head injury and substance
misuse, low non-verbal ability, childhood hyperactivity-impulsivity
and inattention, impairments in empathy and social cognition,
current anxiety and depression, and-often as a default and catch-all
explanation-developmental dyslexia.
15. Some of these obstacles can be identified
by "taking a history". There is something to be said
for standardising the form of a structured or semi-structured
interview, with both closed and open questions. Because of the
great sensitivity of some of the problems that might arise, these
interviews should be conducted only by trained staff in private.
With the exception of some speakers of English as a second language,
the slowest learners are likely to be people with multiple and
often devastating experience of disadvantage.
16. Other obstacles can be identified through
the use of psychometric tests such as tests of non-verbal ability,
auditory-verbal working memory, ability to sustain attention,
and tests of insight into other people's feelings and intentions.
Some tests of this kind are short and highly engaging for the
test-taker, and can be used by the tutor without arduous training.
17. Psychometric tests of these kinds may
be more encouraging when they indicate what learners can do than
when they reveal—to the tutor if not to the learner—impairments
so severe that little progress can be expected. If they are used
at all, they should be interpreted as identifying the difficulty,
not as pigeonholing the learner. A thorough investigation might
explore the functional domains served by all five large-scale
neurocognitive networks: spatial awareness, language, memory and
emotion, working memory and executive function, and face and object
identification [3].
In brief
— Initial basic skills screening can
help to determine whether someone can cope with the reading demands
of a rehabilitation programme; however, test results obtained
during the induction period are often misleading.
— Diagnostic assessment after the induction
period can establish the nature and extent of a learner's reading
skills deficits.
— If reliable data is to be obtained
on induction or later, the timing and location of assessment need
to be both standardised and optimised.
— If the learner appears to be making
little or no progress after six weeks or so, the assessment of
reading skills deficits might need to be followed by an investigation
into impediments to learning.
— Such an investigation might entail
"taking a history" and administering psychometric tests.
— Because "taking a history"
might lead to the disclosure of matters of great sensitivity,
it should follow a standard protocol designed to minimise distress
and be conducted only by trained staff.
— A thorough psychometric investigation
might assess functionality in five domains.
Note 1: In IALS, literacy levels
III, IV and V are considered functional.
Note 2: The original data were expressed
as percentages of the blue-collar and white-collar populations
respectively. They have been re-calculated as percentages of the
total population, using the ratio of blue-collar to white-collar
participants from the 1991 Census.
Source: Table A2.6 "Literacy Level
by Social Class", in Carey, S, Low, S, & Hansbro, J (1997).
Adult Literacy in Britain. London: The Stationery Office,
for the Office for National Statistics.
REFERENCES
Rice, M (2000). The extent and nature of reading
problems in the prison population. Prison Service Journal
(129), 2-10.
Rice, M, & Brooks, G (2004). Developmental
Dyslexia in Adults: a research review. London: National Research
and Development Centre for adult literacy and numeracy.
Rice, M (submitted). A review of Parsons, S
(2002). Basic Skills and Crime: findings from a study of adults
born in 1958 and 1970, published by the Basic Skills Agency.
June 2004
|
