Parental attitudes to children's journeys to school

Children’s journey to school: spatial skills, knowledge and perceptions of the environment

Mary Sissons Joshi, Morag MacLean and Wakefield Carter

Psychology Department Oxford Brookes University

Published as: Sissons Joshi, M., MacLean, M. & Carter, W. (1999) Children’s journey to school: spatial skills, knowledge and perceptions of the environment. British Journal of Developmental Psychology, 17, 125-140.

2

Children’s journey to school: spatial skills, knowledge and perceptions of the environment

Abstract

The growth in accompanied travel to school, particularly by car, has led to

speculation about the cognitive and emotional impact of this change on child

development. Spatial skills, knowledge of the environment, and perceptions of the

environment were assessed in 93 children aged between 7 and 12 years. Children

who were accompanied to school performed as well as their unaccompanied peers on

spatial ability tests and showed no greater concern with stranger danger. However,

they showed a greater tendency to cite traffic danger in their responses, and a greater

knowledge of the environment as indicated by use of landmarks in their drawings of

their locality. Children who had more freedom to travel without adults on non-school

journeys also showed a greater use of landmarks. Mode of transport had no effect on

the study’s measures. These results are discussed with reference to the nature of the

journey to school and to other places.

3

INTRODUCTION

The growth in accompanied travel to school, particularly by car, has been one of the

most pronounced changes in travel behaviour in the UK over the last 20 years.

Hillman, Adams and Whitelegg (1990) have noted how the proportion of 7 to 8 year

old children who travel to school independently declined from 80% in 1970 to 10% in

1990. The phenomenon of accompanied travel to school has several important

consequences. The journey to school is a major source of traffic at peak hours and

adds to congestion, accidents and pollution (Royal Commission on Environmental

Pollution, 1994). Such journeys also cost parents time (Gershuny, 1993) and deprive

children of the opportunity for regular exercise (Armstrong, 1993). A provocative

idea in this area is Hillman’s assertion that a prolonged period of escort to school and

other destinations is likely to hamper development of children’s spatial skills, limit

their knowledge of the environment and damage their growing independence

(Hillman et al., 1990; Hillman, 1993). The assumed deleterious connection between

accompaniment and child development is frequently cited as fact not only in the

media but also in government policy statements such as ‘Developing a Strategy for

Walking’ (Department of Transport, 1996a) and ‘The National Cycling Strategy’

(Department of Transport, 1996b).

Hillman et al.’s (1990) assertions, by their own account, mostly derive from an

unspecified “reading of the child development literature” (p. 81), but some support for

their ideas can be found in the psychological literature. Following their classic study,

Held and Hein (1963) concluded that in kittens and possibly humans, a relationship

existed between locomotor experience and the understanding of spatial relations.

While some authors have suggested that the development of spatial skills depends

4

more on tracking skills than on independent movement (Walk, 1981; Acredolo,

Adams & Goodwin, 1984), others have maintained an emphasis on the importance of

movement. Hazen (1982), for example, showed that two to three year olds with an

active style of exploration were better able to reverse a taught route through a

playhouse than were children with a more passive style of exploration. In a similar

vein, Campos, Benson and Rudy (1986) showed that infants with extensive

experience of moving around in baby-walkers had a superior understanding of their

environment than did other infants (as measured by performance on object

permanence tasks).

There have been a variety of studies in this domain with older children, commencing

with Hart’s (1979) pioneering research into children’s place knowledge, use and

preference. In his widely cited qualitative study, children aged between five and ten

years were required to create small-scale models of their town in New England. Hart

found that even children as young as five years were able to produce spatially

organised representations of areas familiar to them, and that their opportunities to

experience the “geographic-scale environment” (p.112) had a marked effect on the

organisation of their representations. Length of time spent in the area was associated

with a greater use of roads and landmarks in their models. Wapner, Kaplan and

Ciottone (1981) also showed that children’s representations of the area in which they

lived were more related to the amount of time they had lived there than to their age.

Studying the influence of active versus passive experience of the environment,

Cornell and Hay (1984) found that route learning was superior amongst five and eight

year old children who had previously walked the route with a guide rather than seen it

on a slide or video presentation. Nerlove, Munroe and Munroe (1971) studied five to

5

eight year old children in rural Kenya and found that children whose daily tasks (such

as minding cattle or collecting water) took them further from home, were more skilful

at spatial tasks than were their less-travelled age-matched counterparts. In an

experimental study, Poag, Cohen and Weatherford (1983) found that five year old

children who roamed freely in a classroom area made more accurate estimations of

distances than those who were directed by an adult in the same area. Furthermore,

Matthews (1987) found correlations between children’s experience of their local

environment (as measured by the distance from home children were permitted to

travel unaccompanied) and the amount of information about their locality they

included in their maps. Thus it can be seen that some psychologists have taken a

position similar to Hillman et al.’s (1990) and argued that amount and type of

environmental experience will be crucial in the development of spatial representations

of the environment.

Focusing the discussion more specifically on travel in everyday life, Hart (1981) has

suggested that mode of travel in an environment will be important, and that if physical

movement is relevant to how children learn routes, children will represent routes

more effectively under some conditions than others (e.g. walking to school vs. taking

the bus). Lee (1963) investigated the effects of mode of transport to school amongst

six to seven year old children in rural Devon but in the context of emotional

development rather than the development of spatial skills or knowledge of the

environment. He found that children with long journeys to school (either by bus or on

foot) scored poorly on an index of emotional adjustment but no details are given of

accompaniment.

6

In view of the ever-increasing numbers of children whose mode of transport is a

passive one (in the sense of being driven or accompanied on foot), a contemporary

study is needed to ascertain the effects this may have on the development of spatial

skills. It will be important to study the journey to school since it is both daily and at

this age universal, but of course it is not the only local journey children make. They

also go to shops and play-areas and visit friends (Department of Transport, 1996c).

Children can make such journeys by themselves, with other children, or with adults,

and these travel experiences may also have a bearing on their development.

The discussion above, and indeed the psychological literature, have focused on the

relationship between environmental experience and cognitive development. However

it will also be important to establish whether children’s mode of travel has an impact

on their emotional response to the environment. A variety of studies have shown that

contemporary parents limit their children’s freedoms because of traffic danger and

stranger danger (Hillman et al. 1990; Joshi & MacLean, 1995; Valentine G, 1996a,

1996b; Joshi, MacLean & Carter, 1997). Considerably less research has been carried

out on whether children themselves experience the environment in these terms.

Hillman (1993) fears that over-anxious parents have produced a generation of

children who not only have limited independence but are also fearful of their

environment. In Valentine’s (1997) work children aged 8 to 11 reported that stranger

danger warnings and rules had been “drummed into them” by their parents at an early

age, but that they thought themselves to be “competent at negotiating their own safety

in public space” (p. 77). Valentine’s qualitative data, while rich and informative,

emerge from group discussions. Such a methodology does not permit the mapping of

individual children’s images onto those of their parents.

7

Generally, children’s needs and requirements of the environment have not been

sufficiently understood, acknowledged or planned for (Spencer & Blades, 1985;

Roberts, 1996). This study focuses on primary school children since it is at this age

that children begin to travel unaccompanied on the school journey (Hillman et al.,

1990; Bradshaw, 1995; Joshi & MacLean, 1995; Joshi, MacLean & Carter, 1997).

Comparisons will be made of children’s spatial ability, knowledge of the

environment, and perceptions of the environment with respect to (1) accompaniment

on the journey to school, (2) mode of travel to school, and (3) other freedoms. In

addition, for those children who are accompanied to school, the relationship between

parents’ and children’s fears about traffic and stranger danger will be investigated.

METHOD

Subjects

Two County Primary schools agreed to take part in the study. The schools were

located in residential areas in towns with populations of around 20,000. Parents of

children aged between 7 and 12 years were invited to complete a questionnaire

concerning their child’s journey to school (patterns of accompaniment and mode of

travel); their reasons for accompaniment (where relevant); and the extent to which the

child is accompanied on out-of-school journeys/activities. The completed

questionnaires (returned at a rate of 67%) were used to select a sample with

approximately equal numbers of children in each of three categories of journey to

school (car journey with an adult, walking with an adult, and walking either alone or

with other children). Travel by bicycle or public transport was extremely rare for

8

children in these schools. The groups were balanced as far as possible for age and

gender (see Table 1).

In an attempt to hold distance constant, children who lived beyond two and a half

miles from school were excluded from the sample. Ninety-four percent of the

children had lived in the area for more than two years, and the remaining six percent

for at least a year. Eighty-two percent of the sample lived in households with an

annual income between £10,000 and £30,000 (parents having been asked to describe

their household income per annum on a four point scale: up to £10,000; £10,000 -

£20,000; £20,000 - £30,000 and over £30,000).

__________________________

Table 1 around here

__________________________

Materials and Procedure

Hillman (1993) has asserted that prolonged escort on the school journey is likely to

hamper child development and his writings suggest that he is interested in the most

abstract and the most specific of children’s abilities. Thus the measures selected for

this study reflect the two ends of this continuum. Spatial ability was measured in

terms of performance on abstract tasks in close-body space, and knowledge of the

environment was measured in terms of representations of the local area.

Spatial Ability

9

The measures included two tests likely to be pertinent to use of the environment

(locating North and directional orientation) and, from the many tests of spatial ability,

two tests of general spatial ability (a test of mental rotation and the Block Design sub-

test from the Weschler Intelligence Scale for Children).

Locating North

The first of the specific measures was a test of knowledge of compass direction,

namely ability to locate North (q.v. Harris, 1981). Success on this task demands an

awareness of one’s location within a building, and the orientation of the building to

the compass points. Compass knowledge is covered in the primary geography

National Curriculum, and indeed both schools had the points of the compass painted

on the playground (but not within sight of the test room). Each child was presented

with an unmarked dial with a moveable arrow attached to its centre. The desk in the

test room was arranged so that the children faced West and the arrow was also

pointing West. Children were asked ‘Which way do you think North is from here?’

and required to answer by moving the arrow to indicate North. Responses were

scored on a scale ranging from 0 (correct to within 10 degrees of due North) to 18

(due South).

Map orientation (left/right turns)

The second test of specific spatial ability assessed children’s mastery of conventional

directional terms (left and right) and their ability to co-ordinate this with their spatial

orientation. This test was adapted from the Road Map Test of Direction Sense

(Money, Alexander & Walker, 1965). Children were asked to move a figure along a

10

designated path through a schematic map and say whether the figure had to turn left

or right at each of the 24 angled turns.

Mental Rotation Task

A forced choice mental rotation task was added to the battery in order to include a test

of spatial ability which does not depend on left/right labelling. This task was

modelled on Tapley and Bryden’s (1977) rotation test for primary school children.

The children were presented with a drawing of a shape and a drawing of its outline.

The outlines could be rotations of the stimulus image (through 30, 60, 90, 120 or 150

degrees), and were either images or mirror images of the original. Children were

asked to decide whether the shape (“object”) would fit in the outline (“hole”), and to

touch one of two response cards to indicate yes or no. There were five training trials

with instruction on mental rotation and corrective feedback followed by 20 trials

without feedback. Each response was scored as correct or incorrect.

Block Design

The Block Design sub-test of the Wechsler Intelligence Scale for Children was used

in addition to the tasks described above to provide a standardised test of spatial

ability.

Knowledge of environment

The wide variety of techniques that have been used to elicit young children’s

knowledge of the local environment attests to the difficulties of assessing this kind of

knowledge. We rejected measuring children’s knowledge of their environment by

interviewing them as they travelled home on the grounds of impracticality and

11

susceptibility to interviewer bias. Further, verbal descriptions of the environment

given by young children, whether in field or laboratory, can be too brief or too

ambiguous to be useful (Hughes & Grieve, 1980). Model building as a measure of

children’s representations of their environment has been criticised on the grounds that

performance reflects children’s understanding of models as much as their knowledge

of their environment (Siegel & Cousins, 1985). As Spencer and Darvizeh (1981 have

commented, the route maps of children aged three to four years can sometimes consist

of little more than a wavy line with a building at each end, but children’s abilities in

this domain do improve considerably with age (Hart, 1979; Waller, 1986a, 1986b).

Furthermore, while recognising problems surrounding the interpretation of children’s

maps (Downs, 1985: Pocock, 1976), Matthews (1992) notes that adding structure to

the tasks improves performance and his research indicates that features of maps

drawn under free recall conditions are associated with aspects of knowledge of the

environment. Indeed Matthews’ (1992) view is that “ .. children’s maps contain much

beneath the surface, they tell us not only about ... problem-solving in general, but also

about children themselves and what matters to their world” (p. 112). For these

reasons a free-recall map drawing task was included in the present study as a measure

of children’s environmental knowledge.

Map drawing task

Children were provided with an A2 sheet of paper with an outline of their school

buildings printed in the centre of the sheet. In line with Mayall’s (1993) emphasis on

the importance of the space between home and school to children, they were asked to

draw a map of the area around their school, and also asked to “show where you live,

where you play, and the other places you go”. In addition they were told that they

12

could write labels on their maps if they wished. Following Hart (1979) and Matthews

(1992), maps were scored for three features: numbers of landmarks depicted (e.g.

other schools, Tescos and other shops, car parks and bridges indicated by labels or

pictures), the number of closed loops made by roads or paths, and numbers of

different types of pictures drawn, (e.g. houses, trees, swings and slides; with the

exclusion of features drawn within the school grounds). These three categories

allowed the children’s scores to reflect variation in both type and amount of

information about the locality included in their maps. The coder was unaware of how

individual children travelled to school.

Vocabulary

In order to include a standardised non-spatial measure of intellectual ability,

children’s verbal skills were tested using the British Picture Vocabulary Scale

(B.P.V.S.) (Dunn & Dunn, 1982).

Children’s perceptions of the environment

To assess more general perceptions of the environment, children were invited to

respond to four sets of drawings depicting hypothetical routes home and one set of

photographs depicting possible play areas. Specially prepared drawings were used in

order to effect systematic variation in the scenes depicted, and photographs were used

to provide the children with engaging material to elicit their responses to play areas.

The choice of this methodology follows a long-established tradition of using visual

stimuli to elicit children’s evaluations of the environment (Axia, Peron & Baroni,

1991). The children were asked to imagine that they were walking home from school

on their own without any other children. For each of the four sets of drawings they

13

were asked to specify their most and least preferred routes home and to give

justifications for their decisions. The choice points were: 1) a park versus a road, 2)

five pictures of a park systematically varying the numbers and types of people

present, 3) five pictures of a residential street systematically varying numbers and

types of people present, and 4) five pictures of a trunk road systematically varying

amount and type of traffic (see Figures 1 and 2 for illustration of stimulus materials).

Finally, to broaden the assessment of children’s perceptions to include their

experiences of the environment beyond the school journey, they were shown six

colour photographs depicting possible play areas. They were asked to choose the area

in which they would most like to play, the area in which they would least like to play,

and to justify their choices. The photographs were taken in localities that were not the

children’s own and showed: 1) a path through a wood, 2) a bridge over a river

leading to a meadow, 3) a rural track with a parked van, 4) waste ground behind

houses, 5) a towpath by a wooded canal, and 6) a towpath by an industrial canal.

There were no people in any of the six photographs. Children’s responses to the

stimuli were recorded and the transcriptions were coded for reference to traffic danger

and stranger danger.

__________________________

Figures 1 and 2 around here

__________________________

Adult questionnaire

In addition to questions detailing mode of transport to school, all parents indicated the

conditions under which they permitted their children to undertake four other types of

14

journey/activity : going to the park/playground; playing in the street; going to shops

within half a mile; going to more distant shops. The parents could select from four

response options : never; only with an adult; only with other children; alone. The

responses to these four questions were summed to create a composite ‘freedoms’

score (with a possible range of 4-16). Parents who accompanied their child to school

were asked to indicate their reasons for so doing by checking a list of alternatives

which included traffic danger and stranger danger.

RESULTS

The sample comprised 93 children (48 boys, 45 girls) with a mean age of 9 years

(ranging from 7 years 3 months to 11 years 4 months). Amongst the children 31.2%

travelled to school by car, 34.4% walked with an adult and 34.4% walked alone or

with other children. Of those unaccompanied by an adult, 81% travelled alone or

with children of a similar age and the remainder travelled with older children. Each

child’s journey to school was classified as accompanied/unaccompanied by an adult,

and also classified as being by car/on foot (see Table 1). Table 2 gives the means and

standard deviations on the measures of spatial ability, map drawing and freedoms.

__________________________

Table 2 around here

__________________________

A series of multiple regressions, using backward elimination, were conducted to

investigate the relationships between the key variables of accompaniment, mode of

travel and freedoms and the outcome measures - spatial ability, map drawing, and

15

children’s perceptions. The regressions also included three additional explanatory

variables : viz. B.P.V.S., age and gender.

__________________________

Table 3 around here

__________________________

Table 3 shows that none of the four measures of spatial ability were predicted by

accompaniment, mode of travel or freedoms. However, three of the four measures

were predicted by age and B.P.V.S. Older children and those with higher vocabulary

scores performed better on the map orientation (left/right turns), mental rotation and

Block Design tasks.

__________________________

Table 4 around here

__________________________

The results from the map drawing task were more complex (see Table 4). Mode of

travel did not predict any of the three measures. Accompaniment and freedom did not

predict variety of picture types or use of closed loops, but did predict children’s use of

landmarks, which was also predicted by age and B.P.V.S. That is to say, being

accompanied on the school journey, greater travel freedoms on other journeys,

increasing age, and higher vocabulary scores, were all related to the depiction a

greater number of landmarks in the map drawing task. Gender was only a significant

contributory factor in the prediction of the use of closed loops, with boys scoring

higher than girls. Children’s use of landmarks was positively correlated with

16

performance on Block Design (r = 0.34, p <0.01) and map orientation (r = 0.28, p

<0.01) but not with locating North or the mental rotation task. Figure 3 shows

examples of maps which scored high and low on landmarks and loops.

__________________________

Figure 3 around here

__________________________

Turning to children’s choice of preferred route home, 82% of the children selected the

park and 18% selected the road, and a logistic regression on this binary variable

demonstrated that this choice was not predicted by accompaniment, mode of travel,

freedoms, or any of the other variables in the study (backward likelihood ratio,

removal probability = 0.10). Amongst the photographs of potential play places the

most favoured place was the path through the wood, chosen by 63% of children, and

was not related to any of the study’s variables (see Table 5). The most disliked place

was the towpath by an industrial canal, nominated by 56% of children and a logistic

regression showed that this locale was particularly disliked by the older children

(Wald = 5.28, df = 1, p = 0.02)

As Table 6 shows, traffic danger was mentioned by 68% of children in the

justifications they offered at the five choice points (four routes home and one play

place) and was predicted in a logistic regression by accompaniment but not by mode

of travel, other freedoms, age, gender or B.P.V.S. Of the children accompanied by an

adult on the journey to school, 75% mentioned traffic danger in comparison to 53% of

the unaccompanied children (Wald = 3.98, df = 1, p<0.05). Stranger danger was

17

mentioned by 31% of children, and a logistic regression demonstrated that this choice

was not predicted by accompaniment, mode of travel, freedoms, or any of the other

variables in the study. In commenting on the pictures and the photographs, features

other than traffic danger and stranger danger were also mentioned by the children.

These included references to ‘people’, ‘children’ and ‘trees’ but with no indication of

whether these features were positively or negatively valued. However, pollution

(references to ‘noise’ and/or ‘smells’) was unequivocally negatively evaluated by the

children and mentioned by 70% of them.

The relationship between child and parent mentions of traffic and stranger danger

were investigated in the sub-set of 61 children accompanied on the journey to school

for whom parental reasons for accompaniment were available. Amongst these parents

(all of whom were mothers), 64% referred to traffic danger and 90% referred to

stranger danger (see Table 6), and parental references did not explain variation in

children’s references to these dangers.

For this same sub-sample of accompanied children, the relationship between parental

reasons for accompanying their children on the journey to school and child freedoms

were investigated in a multiple regression. Freedoms were predicted by child’s age

and parental reference to stranger danger, i.e. more freedom was granted to older

children and to those whose parents were less likely to refer to stranger danger as one

of their reasons for accompanying their child on the journey to school (see Table 7).

Freedoms were not related to the child’s gender or B.P.V.S. or to parental reference to

traffic danger.

18

DISCUSSION

Spatial ability on three of the four measures in the study was higher in older children

and in children with higher vocabulary scores, but was unrelated to our major

variables. Thus the study provides no support for the idea that children’s spatial

ability is disadvantaged by either travelling to school with an adult, or travelling to

school by car, or even disadvantaged by being restricted in other local journeys.

Turning to the map drawing task, the use of landmarks was predicted by

accompaniment and freedoms. The landmarks depicted covered a very wide range of

features in the children’s local environment. These included highly prominent

buildings such as the supermarkets Tescos and Asda; smaller shops likely to be very

interesting to children such as sweet and video shops; sports grounds and recreation

areas; streams and train tracks; and details which may not be significant to adults,

such as gate posts in the road outside school. Whether it was on foot or by car,

children who were accompanied to school used more landmarks in their maps than

did children who were not accompanied. An explanation of this may relate to

Solomon’s (1993) comment that accompanying children on the journey to school may

have some advantages such as promoting parent-child communication and that

“conversations on the move ... are a very good form of communication” (p. 84). Our

data suggest that some of these conversations may well concern the immediate

surroundings, and include the identification and labelling of local landmarks.

Children who had greater freedom to travel without an adult on non-school journeys

also used more landmarks in their maps than children with less freedom. In order to

explain this apparently paradoxical finding it must be noted that while being

19

accompanied on the school journey is related to children’s other travel freedoms (r =

0.60, df = 90, p<0.001), it is by no means perfectly related. It is likely that

accompaniment on the school journey is determined not only by parents’ attitudes to

children’s licences but also by particular factors such as the specific route involved

and the timing of other household journeys (e.g. taking other children to school and

travelling to work) (Gershuny, 1993; Bradshaw, 1995; Department of Transport,

1996c). The freedom measure on the other hand is a composite of four types of

journey and may be a better indication of children’s licences than the school journey.

Age and B.P.V.S. predicted landmark use such that older children and children with

higher vocabulary scores produced more landmarks, supporting Liben’s (1991)

suggestion that “children’s growing linguistic facility … permits a better coding of

environmental information” (p. 258).

At all ages boys showed a greater use of closed loops in their maps than girls (see

Table 4, β = -0.268, p <0.01). Use of loops involved drawing networks of paths and

roads linking environmental features such as football fields and shops. There was a

similar trend for boys to use more landmarks than girls although this did not achieve

statistical significance (see Table 4, β = -0.162, p = 0.08). One possible interpretation

of the boys’ apparent superiority in two of the three map measures is a cognitive one.

Kitchen’s (1996) review of sex differences in spatial knowledge, cognitive mapping

and geographical knowledge notes that many studies find no reliable differences

between boys and girls in cognitive mapping tasks but that where differences emerge,

boys tend to do better than girls. Sex differences have often been explained with

reference to innate predispositions (Linn & Petersen, 1985). Sex differences in

cognitive maps (of which our map drawing task is an example) have also been

20

ascribed to motivation (Baenninger & Elenteny, 1997) and to the strategies males use.

For example, Miller & Santoni (1986) found that males made more use of Euclidean

properties in their maps whereas females tended to concentrate on surface features.

This finding helps us interpret the boys’ superiority on closed loops, an example of

focusing on Euclidean properties. However, another interpretation of the boys’

advantage is that they spend more time beyond the home than do girls, i.e. boys are

more likely to take up freedoms granted by their parents. Armstrong et al. (1990)

have shown boys to be more active than girls from the age of 11 years and national

travel data suggest that boys aged 5 to 10 years cover more miles cycling per year

than do same aged girls (Department of Transport, 1996c). Matthews (1987) found

that in suburban Coventry, boys of 6 to 11 “often roamed more than twice as far as

girls of their own volition and most continued to enjoy the advantages of broader

place experience for the remainder of their middle childhood” (p.25).

Sixty eight percent of children mentioned traffic danger when justifying their

preferred route home and places to play. These mentions were predicted by

accompaniment on the school journey but not by mode of travel, freedoms, age or

gender . That is to say, two-thirds of children who were accompanied on the school

journey mentioned traffic danger compared with only half of the unaccompanied

children. As suggested above, the school journey may provide daily conversational

opportunities for parents including reminders of traffic danger. Only 31% of children

mentioned stranger danger and such mentions were not predicted by accompaniment,

mode of travel, freedoms, age or gender.

21

It was only the parents of the 61 children who were accompanied on the school

journey who were given the opportunity to express concerns about their child’s

environment. Sixty-four percent mentioned traffic danger and 91% mentioned

stranger danger. Within families, no statistically significant relationships were found

between parent and child mentions of traffic and stranger danger. Both the differing

levels of concern in respect of stranger danger, and the lack of correlation between

parent and child dyads, may in part be task dependent. The parents’ mentions were

responses to a specific question about their reasons for accompanying their child to

school, and elicited via a check-list. In contrast, the children’s mentions were

responses to a series of questions about hypothetical situations and elicited via T.A.T.

style stimuli.

A further explanation of the children’s comparatively infrequent mentions of stranger

danger lies within the probable content of their conversations with their parents.

Parents are extremely concerned about stranger danger (Hillman et al., 1990; Joshi &

MacLean, 1995, Joshi, MacLean & Carter, 1997; Valentine, 1996a, 1996b), and in

the current study parental concern with stranger danger was predictive of general

travel freedoms granted to children, as was the child’s age. However the importance

to parents of stranger danger does not mean that parents’ everyday conversations with

their children are dominated by discussions of stranger danger. In contrast, it is

highly likely that traffic danger is a frequent subject of discussion as every journey

provides ample opportunities for reminders and instruction in this area.

The point of departure for our study was Hillman et al.’s (1990) conviction that a

prolonged period of escort hampers children’s development of spatial skills as well as

22

depriving them of the opportunity to develop local knowledge, take initiative, acquire

practical coping skills and develop self-esteem. We did not attempt to study the

broadest of these competencies but our study has suggested that neither

accompaniment on the journey to school nor mode of travel to school affect spatial

abilities measured in decontextualised tasks. We did not find stranger danger to be

more common amongst accompanied children, nor did we find such an anxiety being

transmitted from individual parent to child as there was no association between

parental and child mentions of stranger danger. Children as a group showed far less

concern with stranger danger than did their parents, but showed as much concern

about traffic danger. Being accompanied to school was associated with a greater

knowledge of the environment, as was being granted greater freedom to travel without

an adult on non-school journeys. Unambiguous causal conclusions cannot be drawn

from correlational data such as these, and an experimental design whereby children

are assigned to transportation mode is not possible in this area of study. While

knowledge of the environment is important in its own right, empirical research has yet

to settle the issue of whether the acquisition of such knowledge at an early age

confers any developmental advantage, and if so, in what sphere. Longitudinal

research is of course required to investigate the causal relationships between variables

rather than simply deal in correlation as the present study has done. It will also be

necessary to investigate the effects of prolonged accompaniment and car dependence

not only on children but also on adolescents, and in particular on their perceptions and

use of the environment.

Public debate and transport specialists have focused on the journey to school as it is

the largest single category of journey that children make (Department of Transport,

23

1996c) and accompaniment by car contributes to traffic congestion, accident rates and

pollution (Royal Commission on Environmental Pollution, 1994). However, by its

routine nature, the journey to school may not be as crucial to child development as

Hillman et al. (1990) proposed. Indeed our data clearly indicate that it is the freedom

to travel without an adult on journeys other than the school journey which is

predictive of knowledge of the environment.

24

REFERENCES Acredolo, L.P, Adams, A., & Goodwyn, S.W. (1984) The role of self-produced

movement and visual tracking in infant spatial orientation. Journal of Experimental Child Psychology, 38, 312-327.

Armstrong, N. (1993) Independent mobility and children’s physical development. In

M. Hillman (Ed.) Children, Transport and the Quality of Life, London: Policy Studies Institute.

Armstrong, N., Balding, J., Gentle, P. & Kirby, B. (1990) Patterns of physical activity

among 11 to 16 year old British children. British Medical Journal, 301, 203-205.

Axia, G., Peron, E.M. & Baroni, M.R. (1991) Environmental assessment across the

life span. In T. Garling & G.W. Evans (Eds.) Environment, Cognition, and Action: An Integrated Approach, New York: Oxford University Press.

Bradshaw, R. (1995) Why do parents drive their children to school? Traffic

Engineering and Control, January 1995, 16-19. Baenninger, M. & Elenteny, K. (1997) Which way do we go? American Scientist,

85(6). Campos, J.J., Benson, J., & Rudy, L. (1986) The role of self-produced locomotion in

spatial behavior. Poster paper presented at the meeting of the International Conference for Infant Studies, Beverly Hills, CA. (cited in M. Cole & S.R. Cole The Development of Children, 2nd edition, 1993, New York: Scientific American Books.

Cornell, E.H. & Hay, D.H. (1984) Children’s acquisition of a route via different

media. Environment and Behavior, 16 (5), 627-641. Department of Transport (1996a) Developing a Strategy for Walking. London:

Department of Transport. Department of Transport (1996b) The National Cycling Strategy. London:

Department of Transport. Department of Transport (1996c) National Travel Survey 1993/95. London: HMSO. Downs, R. M. (1985) The representation of space: development in children and in

cartography. In R. Cohen (Ed.) The Development of Spatial Cognition. Hillsdale, New Jersey: Lawrence Erlbaum.

Dunn, L.M. & Dunn, L.M. (1982) British Picture Vocabulary Scale, Windsor: NFER-

Nelson.

25

Gershuny, J. (1993) Escorting children : impact on parental lifestyle. In M. Hillman (Ed.) Children, Transport and the Quality of Life, London: Policy Studies Institute.

Harris, L.J. (1981) Sex-related variations in spatial skill. In L.S. Liben, A.H. Patterson

& N. Newcombe (Eds.) Spatial Representation and Behaviour across the Life Span: Theory and Application. New York: Academic Press.

Hart, R.A. (1979) Children’s experience of place. New York: Irvington. Hart, R.A. (1981) Children’s spatial representation of the landscape: Lessons and

questions from a field study. In L.S. Liben, A.H. Patterson & N. Newcombe (Eds.) Spatial Representation and Behaviour across the Life Span: Theory and Application. New York: Academic Press.

Hazen, N.L. (1982) Spatial exploration and spatial knowledge: individual and

developmental differences in very young children. Child Development, 53, 826-833.

Held, R. & Hein, A. (1963) Movement-produced stimulation and the development of

visually guided behaviours. Journal of Comparative and Physiological Psychology, 56, 872-876.

Hillman, M., Adams, J., & Whitelegg, J. (1990) One False Move ... A Study of

Children’s Independent Mobility, London: Policy Studies Institute. Hillman, M. (1993) One False Move... A study of children’s independent mobility :

an overview of the findings and the issues they raise. In M. Hillman (Ed.) Children, Transport and the Quality of Life, London: Policy Studies Institute.

Hughes, M. & Grieve, R. (1980) On asking children bizarre questions. First

Language, 1, 149-160. Joshi, M.S. & MacLean, M. (1995) Parental attitudes to children’s journeys to school.

World Transport Policy and Practice, 1(4), 29-36. Joshi, M.S., MacLean, M. & Carter,W. (1997) Children’s journeys to school: new

data and further comments. World Transport Policy and Practice, 3(4) in press.

Kitchen, R.M. (1996) Are there sex differences in geographic knowledge and

understanding? The Geographical Journal, 162(3), 273-286. Lee, T.R. (1963) On the relation between the school journey and social and emotional

adjustment in rural infant children. British Journal of Educational Psychology, 27, 100-114.

Liben, L.S. (1991) Environmental cognition through direct and representational

experiences: A life-span perspective. In T. Garling & G.W. Evans (Eds.)

26

Environment, Cognition, and Action: An Integrated Approach, New York: Oxford University Press.

Linn, M.C. & Petersen, A.C. (1985) Emergence and characterization of sex

differences in spatial ability: a meta-analysis. Child Development, 56, 1479-1498.

Matthews, M.H. (1987) Gender, home range and environmental cognition.

Transactions of the Institute of British Geographers, new Series, 12, 43-56. Matthews, M.H. (1992) Making Sense of Place: Children’s Understanding of Large-

Scale Environments. Hemel Hempstead: Harvester Wheatsheaf. Mayall, B. (1993) Keeping healthy at home and school: ‘it’s my body, so it’s my job’.

Sociology of Health and Illness, 15, 464-487. Miller, L.K. & Santoni, V. (1986) Sex differences in spatial abilities: strategic and

experiential correlates. Acta Psychologica, 63, 225-235. Money, J., Alexander, D. & Walker, H.T. (1965) A Standardized Road-Map Test of

Direction Sense. Baltimore: Johns Hopkins Press. Nerlove, S.B., Munroe, R.H. & Munroe, R.L. (1971) Effect of environmental

experience on spatial ability: a replication. Journal of Social Psychology, 84, 3-10.

Poag, C.K., Cohen, R. & Weatherford, D.L. (1983) Spatial representations of young

children: the role of self versus adult directed movement and viewing. Journal Experimental Child Psychology, 35, 172-179.

Pocock, D. (1976) Some characteristics of mental maps: an empirical study.

Transactions of the Institute of British Geographers, New Series, 1, 493-512. Roberts, H. (1996) Intervening to prevent accidents. In B. Gillham & J.A. Thomson

Child safety: problem and prevention from preschool to adolescence. London: Routledge.

Royal Commission on Environmental Pollution (1994) 18th report. Transport and the

Environment. London: HMSO. Siegel, A.W. & Cousins, J.H. (1985) The symbolizing and symbolized child in the

enterprise of cognitive mapping. In R.Cohen (Ed.) The Development of Spatial Cognition. Lawrence Erlbaum : Hillsdale, New Jersey.

Solomon, J. (1993) Escorting: balancing the advantages and the disadvantages. In M.

Hillman (Ed.) Children, Transport and the Quality of Life, London: Policy Studies Institute.

27

Spencer, C. & Blades, M. (1985) Children at risk: Are we underestimating their general environmental competence whilst overestimating their performance? In. T. Garling & J. Valsiner (Eds.) Children within Environments: Toward a Psychology of Accident Prevention, New York: Plenum Press.

Spencer, C. & Darvizeh, Z. (1981) Young children’s descriptions of their local

environment: a comparison of information elicited by recall, recognition and performance techniques of investigation. Environmental Education and Information, 1, 275-284.

Tapley, S.M. & Bryden, M.P. (1977) An investigation of sex differences in spatial

ability: mental rotation of three dimensional objects. Canadian Journal of Psychology, 31, 122-130.

Valentine, G. (1996a) Children should be seen and not heard: The production and

transgression of adults’ public space. Urban Geography, 17, 205-220. Valentine, G. (1996b) Angels and devils: Moral landscapes of childhood.

Environment and Planning D: Society and Space, 14, 581-599. Valentine, G. (1997) “Oh yes I can.” “Oh no you can’t”: children and parents’

understandings of kids’ competence to negotiate public space safely. Antipode, 29, 65-89.

Walk R.D. (1981) Perceptual Development, Pacific Grove, CA: Brooks/Cole. Waller, G. (1986a) The Development of the Child’s Ability to Communicate about

Space. unpublished D.Phil. thesis, The University of Oxford. Waller, G. (1986b) The development of route knowledge: multiple dimensions?

Journal of Environmental Psychology 6(2), 109-119. Wapner, S., Kaplan, B. & Ciottone, R. (1981) Self-world relationships in critical

environmental transitions: Children and beyond. In L.S. Liben, A.H. Patterson & N. Newcombe (Eds.) Spatial Representation and Behaviour across the Life Span: Theory and application. New York: Academic Press.

Wechsler Intelligence Scale for Children - Revised Anglicized edition. (1976) New

York: Psychological Corporation.

28

Table 1 Sample description

Journey to school

By car with adult Walk with adult Walk alone/with children

Gender Boys 16 16 16 Girls 13 16 16 Age Mean 8y 6m 8y 9m 9y 8m Range 7y 3m - 10y 8m 7y 4m - 11y 3m 7y 9m - 11y 4m Distance Mean Home-school s.d.

0.88 miles 0.50

0.57 miles 0.23

0.56 miles 0.15

29

Table 2 Spatial Ability, Map Drawing, and Freedoms: Means and standard deviations by age

Age groups

Age 7

Age 8

Age 9

Age 10+

Spatial ability

Locating North (0-18)

8.44 (5.24)

8.57 (5.37)

7.92 (5.29)

8.65 (4.77)

Map Orientation (0-24)

13.11 (3.46)

14.50 (4.14) 15.44 (3.36) 15.60 (2.80)

Mental Rotation (0-20)

13.72 (4.03)

15.13 (4.42)

16.32 (3.79)

17.05 (2.87)

Block Design

23.72 (13.07)

23.70 (9.56)

29.08 (10.65)

35.10 (11.33)

Map Drawing

Picture types (0-13)

3.17 (2.04) 2.97 (2.68)

3.28 (2.52) 2.50 (2.14)

Loops (0-13)

2.33 (2.38) 1.83 (2.57) 2.24 (2.47) 3.45 (3.17)

Landmarks (0-43)

7.94 (4.09) 7.83 (5.39) 14.32 (11.35) 12.65 (7.79)

Freedoms

Scores (4-16)

6.28 (1.45) 6.89 (1.31) 8.13 (1.62)

8.05 (1.36)

N

18

30

25

20

30

Table 3 Multiple regressions on children’s spatial abilities Independent variables Spatial Ability Locating

North Map Orientation

Mental Rotation

Block Design

Accompaniment β -0.065 -0.071 0.012 -0.088 p 0.54 0.53 0.92 0.38

Mode of travel β 0.124 -0.060 -0.060 0.003 p 0.24 0.57 0.57 0.98

Freedoms β -0.071 0.027 0.031 -0.090 p 0.50 0.81 0.78 0.37

Age β -0.010 0.276 0.310 0.460 p 0.92 <0.01 <0.01 <0.01

B.P.V.S. β -0.105 0.240 0.235 0.297 p 0.32 0.02 0.02 <0.01

Gender β 0.007 -0.058 -0.015 -0.053 p 0.95 0.56 0.88 0.55

Adjusted R2 0.104 0.120 0.261 Bold indicates variables significant beyond the 0.05 level.

31

Table 4 Multiple regressions on children’s map drawing Independent variables Map drawing

Picture Types Loops Landmarks

Accompaniment β -0.087 -0.116 -0.295 p 0.46 0.31 0.02

Mode of travel β -0.188 -0.077 -0.067 p 0.07 0.47 0.54

Freedoms β -0.034 -0.098 0.316 p 0.75 0.39 <0.01

Age β -0.092 0.175 0.333 p 0.40 0.09 <0.01

B.P.V.S. β -0.075 0.032 0.319 p 0.47 0.76 <0.01

Gender β 0.086 -0.268 -0.162 p 0.41 <0.01 0.08

Adjusted R2 0.025 0.075 0.231 Bold indicates variables significant beyond the 0.05 level.

32

Table 5 Locale by percentage of children Locale

“Most liked”

“Most disliked”

Path through wood

63.4%

1.1%

Waste ground near houses 15.1% 1.1%

Bridge and meadow 12.9% 17.2%

Wooded canal and towpath

6.5% 19.4%

Rural track and vehicle 1.1% 5.4%

Industrial canal and towpath

1.1% 55.9%

N children

93

93

33

Table 6 Child and adult concerns

Percent concerned with

Traffic danger Stranger danger

All children (N=93)

67.7%

31.2%

Accompanied(N=61)

75.4%

27.9%

Unaccompanied

(N=32)

53.1%

35.0%

Adults who accompany (N=61)

63.9%

90.9%

34

Table 7 Multiple regression on children’s freedoms _____________________________________________________________________ Variable β t p _____________________________________________________________________ Child’s age 0.246 2.11 <0.05 Parent’s reference to -0.388 - 3.32 <0.01 stranger danger _____________________________________________________________________ Variables not predictive of children’s freedoms : BPVS, gender, parent’s reference to traffic danger (sample size = 61, Adjusted R2 = 0.182).

35

Figures 1 & 2 Examples of materials used to elicit children’s perceptions of the environment

36

Figure 3 Children’s maps of the area around the school