10 Action Research Street Children

8/3/2019 10 Action Research Street Children

1/13

Adapting Visual Methods: Action Research with Kampala Street ChildrenAuthor(s): Lorraine Young and Hazel BarrettSource: Area, Vol. 33, No. 2 (Jun., 2001), pp. 141-152Published by: Blackwell Publishing on behalf of The Royal Geographical Society (with theInstitute of British Geographers)

Stable URL: http://www.jstor.org/stable/20004145Accessed: 10/05/2010 07:38

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=black.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Blackwell Publishing and The Royal Geographical Society (with the Institute of British Geographers) are

collaborating with JSTOR to digitize, preserve and extend access toArea.

http://www.jstor.org
http://www.jstor.org/stable/20004145?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/action/showPublisher?publisherCode=blackhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/20004145?origin=JSTOR-pdf


2/13

Area (2001) 33.2, 141-1 52

Adapting visual methods: action research withKampala street children

Lorraine Young and Hazel BarrettDepartment of Geography and Earth Sciences, Brunel University, Uxbridge UB8 3PH and Geography

Subject Group, Coventry University, Coventry CV1 5FBEmails:Lorraine.Younggbrunel.ac.uk and h.barrett(coventry.ac.uk

Revised manuscript received 1 1 July 2000The paper examines the use of four visual 'action' methods for eliciting information fromstreet children about their interactions with the socio-spatial environment. These methodswere adapted and used to encourage child-led activities and minimize researcher input.The advantages and constraints associated with the successful implementation of visualmethods with children are examined, followed by a brief examination of the 'real'andethical considerations surrounding their use. The paper concludes that visual methodsallow a high level of child-led participation in research, as well as providing a stimulus foreliciting further oral material.Key words: Uganda, child-centred research, street children, visualmethods

IntroductuionFollowing on from Aries' (1962) work, childhood hasbecome recognized as socially, rather than biologically, constructed. Social science research now considers children as active, valuable contributors and

members of society (Holloway and Valentine 2000;James et al. 1998; James and Prout 1990; Jenks1996). Holloway and Valentine (2000) recognizethat there are two important elements of the newsocial studies of childhood that are particularly relevant to geographers. First, if childhood is sociallyconstructed, then children must be considered as'meaning-producing beings' in their own right andsecond, childhood must be a culturally diverseexperience for individual children growing up indifferent societies.

Based on this recognition that children are important social actors, researchers within the geographicaldiscipline are now calling for more research on'children as a neglected social grouping undergoing

various forms of socio-spatial marginalization'(Matthews and Limb 1999, 62). At the same timethere is the need to acknowledge the importance of'multiple childhoods' and the difference inboth their

social and spatial constructions (Matthews and Limb1999). A large amount of substantive and informative research has been undertaken in a Westerncontext on the socio-spatial behaviour of childrenand their marginalization from adult society(Matthews 1992; Sibley 1995; Skelton and Valentine1997; Valentine 1996a 1996b 1997). Much less

work has been done on the 'special position ofexclusion' (Matthews et al. 1999, 135) of children indeveloping countries. The street child population is aparticularly interesting facet of multiple childhoodsand marginalization because they are creating theirown spaces within an inherently adult domain, thatis the city. Furthermore, there is a culmination offactors that affect their spatial presence; they areexcluded from public social and spatial environs dueto their status as minors; and they aremarginalized insociety due to their deviant characteristics andhomeless status. Their public existence on the streetsis, therefore, producing a particular 'street childhood'

with associated 'street child spaces', many of whichare hidden and separate from the adult city.

Although street children are recognized throughout the developing world, they have recentlybecome more prevalent inAfrican cities. This is the

ISSN 0004-0894 ?) Royal Geographical Society (with The Institute of British Geographers) 2001


3/13

142 Young and Barrettresult of the combined effects of socio-economicrestructuring, the HIV/AIDS epidemic and internalconflicts which directly impact on the micro situationof the home and family (Harper and Marcus 2000).The geographies developed by these children toenable them to survive in an adult dominated cityscape are little known or understood by policy

makers and researchers. The aim of the researchproject was, therefore, to develop an understandingof Kampala street children's socio-spatial geographies in relation to their street environments andsurvival mechanisms. This paper examines some ofthe methodological issues associated with researching the socio-spatial relationships and spatialexperiences of these children.

Children and the research processMethodologically there are special concerns whenresearching children as

[A]ppropriate esearch strategies, inbothmethodological and ethical senses, need to be thought throughverycarefully. (Sibley 1991, 270)

Recently, however, traditional social science researchmethods have been denounced as problematicbecause they rarely involve children in the researchprocess. These methods are often based on positivistmethodologies using questionnaire surveys for generating large quantities of statistical data. Baker et al.(1996) point out that such strategies are particularlyinaccurate for developing an understanding of achild's circumstances. Moreover, these methodshave rarely enabled children to have any input intothe research design and process, thus highlightingissues of reflexivity and participation, given the difficulties adults have in engaging with the world ofchildren (James 1990 1991; Sibley 1991 1995).

In response, attempts have been made to makeresearch techniques more child-friendly and participatory. Recently, a diverse range of child-centredmethods has been developed across the socialsciences covering oral, written and visual activities.

These include amongst others:* focus group discussions;* observation;* recall;* drawings;* spider diagrams;* resource mapping; and* seasonal calendars.All these methods have been successfully tested in aparticipatory framework for their appropriateness in

eliciting information with children, with manuals forpractitioners being produced on issues of implementation (Boyden and Ennew 1997; IDS 1996; Johnsonet al. 1994).

These child-centred and child-led participatorymethods are particularly relevant when researchingstreet children, given the unique set of circumstancesthat impinge on the lifeworld of a child living andsurviving in the cityscape. Furthermore, the intense,haphazard nature of street life often results in a wayof living that is alien to themajority of researchers. Itis for this reason that Bemak (1996) calls forresearchers to become 'street researchers' getting toknow the street environment and way of life byentering into a mutual trust relationship with thechildren prior to developing and initiating a researchstrategy. According to Rudestam and Newton(1992) researchers must become more creative andflexible inexploring street child environments. Therefore, it is necessary to adapt visual methods to theparticular lifeworld of the street child.

Not only do the unique circumstances of thestreet child population need to be considered whendevising a research methodology, but also themethods selected need to be adapted for elicitingparticular types of information. In this instance, theaim was to develop an understanding of the streetchild's socio-spatial geographies. Given that theytend to locate in 'hidden' spaces within the city,serious methodological issues concerning ethics andpositionality are also raised and must be addressed.A multi-method approach was therefore employedfor this research, allowing the strengths of individual

methods to compensate for limitations in others.Although observation, discussion, role-play and written techniques were used, visual methods provedto be particularly important for developing gainfulinsight into the street child's urban environment fromthe child's perspective. They introduced a relaxed,fun atmosphere and allowed the children to takecontrol of the process without imposing adult influence. This paper illustrates how four visual methodswere successfully adapted expressly for obtainingsocio-spatial information with street children and

within a highly sensitive ethical situation.

Kampala street childrenKampala, the capital of Uganda was selected as acase study for this research as poverty and socialupheaval associated with civil war and the AIDSepidemic have resulted in a sharp rise in the street


4/13

Action research with Kampala street children 143child population in the last ten years. The identification of the socio-spatial relationships and experiences of street children inKampala is important ininforming policy-makers in Uganda, and other

African cities where street children have become amajor urban and development issue. There areapproximately 300 full-time street children whoboth live and survive daily on the city streets ofKampala.1

As indicated in the introduction, defining childhood, and consequently street childhood, is bothcomplex and somewhat arbitrary, particularly if children are recognized as social actors in their ownright.Within the African continent, Ennew (1996)illustrates the range of diverse and multiple experiences that come under the 'umbrella' of childhood.Similarly,within Uganda, children experience a diversity of lifeworlds although these are impinged uponby the legal and constitutional definition of a child.Therefore, although there are numerous childhoodexperiences, each one is impacted upon by theframework set out in The Children Statute (1997)

which draws on the notion of children needing careand protection as outlined in the United Nations'Convention on the Rights of the Child, an essentially

Western construction (Boyden 1991). Therefore, asHolloway and Valentine note:

recognition of children's agency does not necessarilylead to a rejection of an appreciation of the ways inwhich their lives are shaped by forces beyond thecontrol of individual hildren (HollowayandValentine2000, 6).

Given this, the children who participated in this studywere aged between nine and 17 years old.2 Mosthad little or no contact with parents or guardians,having left home principally because of the microeffects of mistreatment, parental death or poverty.They therefore had to create their own survivalopportunities and coping mechanisms. Furthermore,the street children were an eclectic mix of individualscoming from different background situations andwith various mother tongues. Their spoken language,Luyaaye, was a combination of words and phrasesadopted from various Ugandan languages and dialects.When coupled with the fact that many of thechildren in the sample were illiterate, the use ofvisual methods was particularly important forencouraging free expression. Oral and written communication was limited formany of the children inthe sample, as street life resulted in the adaptationof their mother tongue. Visual representation is

universal and could be used as a catalyst for elicitingoral description and information.

In this paper an exploration is conducted into theuse of four visual methods with street children inKampala:1mental and depot3 maps;2 thematic and non-thematic drawings;3 daily time lines; and4 photo diaries.Each visual method is described before the advantages and constraints are discussed, and then theanalytic and catalytic properties of each method areinvestigated. The final section of the paper focuseson the practicalities of conducting research withstreet children highlighting the benefits of participatory visual methods.

Introducing isualmethodsVisual methods have, formany decades, been usedin child development research focussing on spatialcognition and environmental psychology. In testingPiagetian stages of child development, the visualrepresentation of space as a diagram or plan hasbeen a popular method used to demonstrate children's spatial awareness (for example Matthews1984). Researchers have effectively used mental

maps and aerial photographs to show the spatialdevelopment of children by age and gender (Blaut1997; Golledge et al. 1992; Matthews 1984 1987).Similar techniques have also been used to demonstrate the cross-cultural similarities of the mappingabilities of young children (Blades et al. 1998).However, the visual methods used are very muchresearcher directed and controlled. Thus, as Ennew(1996) points out, visual methods applied in this

way inevitably support adultist interpretations ofchildhood. Within this cognitive tradition childrenbecome the objects of the research rather than partof it and often children's interpretation of theirimages are missed by the researcher. This is particularly important in conducting cross-cultural work, as

visuality is a culturally diverse experience (Boydenand Ennew 1997; Swart 1990). By contrast, therecent sociological interest in children as socialactors, uses visual methods to gain insights into thecontext of a child's lived experience (for exampleBeazley 1997). Within this discourse the control andimplementation of research methods are passed


5/13

144 Young and Barrett

from adults to children, allowing them 'to constructaccounts of their lives in their own terms' (Hollowayand Valentine 2000, 8). In this way child-centredvisual methods avoid the adultist assumptions ofthe cognitive school and facilitate research 'withchildren' rather than research 'about children'. Childcentred visual methods can therefore illuminate children's geographies, by embracing and celebratingdifferent experiences of childhood.

For this research a series of visual 'action' activitieswere devised and adapted, which were fun, childcentred and gave the street children involved a largemeasure of ownership of the exercise. They werethen used to elicit information on the street child'suse of the street on a daily basis. Although the four

methods were experimental in nature, some hadbeen used before with street children (for exampleBaker et al. 1 96; Swart 1 90). The experimentation,however, was not just related to the populationunder study but also to their use in eliciting sociospatial data and as a catalyst for oral description aform of basic analysis by the children themselves.

Mental maps and 'depot' mapsThis method was used in a variety of settings atdifferent times. Although 22 children took part, onlysmall groups were gathered together at any one timeto discuss the subject of maps. Once itwas clear thateverybody understood the concept theywere askedto draw their own mental maps of Kampala showingthe places that they usually visited during the day,and depot maps illustrating all the depots where theyknew street children frequented. Inorder to do thiseach child was given a piece of plain paper and apencil already equipped with an eraser should theywish to change their image. The maps were used tofacilitate explanation and discussion with each child,concerning what they had drawn and why they haddrawn it.

The advantages of thismethod for the understanding of children's daily lives were three-fold. First, as

most street children do not attend even informaleducation classes they are limited in their opportunities to use paper and pencils and, as the activitywas action-based rather than discursive and drawingcapabilities were not essential, the children found theactivity fun and others often asked if they could joinin.Second, as such an activity is not spontaneous likeconversational interviews, the children were able totake their time and make amendments to the finished product, thus providing the truest represen

i-- frsc

- OUJ

Figure I Map drawn by a 12-year old boy depicting theplaces in Kampala which 'are important to him'

tation possible. Finally, the maps themselves wereuseful tools in eliciting information about the dailylife of street children as they were keen to talk about

what they had drawn and provide details of whereeach place they had marked was and why itwas animportant place to them. The maps acted as prompts

because often the minds of the street childrenwandered owing to the effects of sniffing fuel. Fur

ther, the spatial representations produced were alsovaluable for examining the 'place' of street childrenin Kampala with many children concentrating their

efforts on the 'down-town' area to the south of thecity. This is generally the busiest and least-regulatedarea containing the markets and taxi parks. Forexample, Figure I is a map drawn by a 12-year oldboy which shows the Old Taxi Park where he sleeps,

Owino Market where he sells vegetables, as well assportsgrounds and video clubs where he spends hisleisure time.

It was often quite difficult for the children toposition places accurately on their maps. On someoccasions the 'discussion' between children on theirmaps became lengthy as some children found itdifficult to grasp the concept. This, however, wasgenerally the result of the effects of fuel inhalation.


6/13

Action research with Kampala street children 145

Copying was also present in these activities althoughnot rife.However, what tended to be copied werenot the actual places drawn but the style of drawing.As these children are out of school their knowledgeof mapping is not well developed. They wouldtherefore discuss how to draw the map beforecreating their own representations. This meant thatmap styles were limited and places were oftenrepresented as circles. However, as the detail wasthe important aspect, itwas not felt that comparingstyles was detracting from the information produced.In hindsight, and time allowing, individual drawingsessions would have eliminated such comparisonsbut reduced enthusiasm generated by initial discussions, and may have returned the situation to adultrather than child-led.

The mental and depot maps produced imagesthat were striking and immediately highlighted theindividuality amongst the participating children. Inparticular the variation in work and leisure spacebecame apparent. The coding of places representedon the maps allowed analysis of the spaces used bydifferent age and gender groups. For example, themental maps demonstrated that, as would beexpected, older boys utilize a greater spatial areawithin the city compared to younger boys and girls.The analysis of the depot maps, when compared tothe results of later focus group discussions, emphasized that there was a core of main depots used bythe children, with other subsidiary sites used lessfrequently. Furthermore, itwas revealed that it is theleast well-established depots that are subject tochange. The visual representations also facilitateddiscussion with the children to explain the importance of the places included in theirmaps.

Thematic and non-thematic drawingsAgain this activity was undertaken in a variety ofplaces and at different times with 23 children. Theactivity was split into two sessions comprising threedrawings each and although some children participated in both, this was not a requirement. This wasdone because of the limited attention span of streetchildren and itwas felt in the pilot study that threedrawings were ideal.Children were each given paperand a pencil and asked to draw a series of pictures.The format varied between formal thematic, informalthematic and non-thematic representations based onaspects of the children's daily lives. Each drawing

was given a title arrived at through piloting procedures and discussions with children. Modifications

were made so that correct interpretations of the titleswould be arrived at.

As with map-making, drawing pictures allowed thechildren to freely express themselves and to thinkabout what they wished to portray. Further, theywere not inhibited about their drawing capabilitieswhich often occurs in more literate children whomay develop a sense of artistic inadequacy. Theimages produced were also useful tools in elicitingdiscussion with individual children as it provided afocus away from the researcher. The children themselves were incontrol of producing the pictures andtherefore, when asked about what they had drawn,the majority talked freely increasing the quality of theinformation gathered.

Poorer images were constructed than those thatwould have arisen from a group of similar-agedschool-going children, due to lack of practice. However, as the quality of the drawing was not the issueunder examination, thiswas not felt to compromisethe results.When drawing formal thematic imagesbased on 'safe' and 'dangerous' places, children occasionally 'mixed up' the pictures because they couldnot remember which column was for each category.This highlighted the importance of the children's subsequent explanations. Often ambiguity occurred and it

was impossible to determine whether the picture represented a safe or a dangerous place. Itwas onlythrough listening to the oral justifications provided thatthe correct understanding could be gained. Althoughseparate papers could have been used and the activitysplit into two, this would not have allowed for children's work to be compared norwould ithave elucidated initialmental conjecture over the meaning ofsafety and danger in the context of space.

Several themes, based on the daily activitiesundertaken by street children, emerged from theanalysis of the drawings. Figures 2, 3 and 4 aretypical of the activities highlighted by street children.

The results suggest that: leisure; work activities,especially collecting rubbish, scrap and carrying luggage; eating; stealing; and sleeping are importantdaily activities for street children. Itwas, however,often difficult to correctly identify the images on thedrawings and to identify themost important features(see also Ennew 1994; Swart 1990). Thus, the pictures were used to prompt more detailed oral infor

mation. This helped elucidate the images accuratelyand the combination of visual and oral methodsresulted in a much richer data set than could havebeen obtained from pictures or discussion alone. Thepictures, therefore, helped overcome problems


7/13


"This isme selling onions in Owino [market] that Ipicked very early in themoming when the trucks wereoff-loading". (D aged 13 years)

Figure 2 Thematic drawing by a 13-year old boy of'working in the market'

"I take rubbish to thedustbin and theypayme" (M aged 9 years)Figure 3 Thematic drawing by a 9-year old boy of'headloading rubbish to a skip'

associated with interview situations, giving control ofthe situation to the children. Lying was reduced asthe children talked freely and the researcher wascareful not to ask direct questions which are knownto evoke fear (Ennew 1994).

Daily time linesThe construction of daily time lineswas elicited as agroup activity. Initially itwas explained to the wholegroup that this particular research taskwas aimed atdiscovering what they did at different times of theday or night. Itwas explained that this would beconstructed using a graph made up of picturesrepresenting different activities. Nineteen childrenthen opted to take partwith three joining in later.A

J)"I have stolen a bag.from this woman and then I havehidden with the bag in the skip" (L aged 10 years)

Figure 4 Thematic drawing by a 10-year old boy of'snatching a bag in the city centre'

discussion ensued on what kinds of activities streetchildren did and what should be included for representation. A listof 11 activities was devised and thechildren devised symbols to represent each one.Often heated discussions arose when someone gottoo artistic and after negotiation the symbol was

modified and simplified until each category had amatching symbol that everyone was happy with.Each child then took turns at creating a line on thegraph representing their typical day and discussingwhere they carried out each of their activities.

This activity in particular, highlighted how themore participatory an activity becomes the moreenthusiastic are the participants. The fun nature ofdesigning the symbols and categories persuadedothers to join in. Furthermore, the initial discussioncould also be highlighted as important for pre-activityanalysis in that the children coded their activitiesthrough the process of designing symbolic representation. There was also an in-built cross-checking

mechanism in that children questioned each otheron what symbols they had drawn or shouted out theactivities of their friends. This was particularly usefulfor de-sensitizing taboo subjects such as stealing. Ifthe children were discussing another's involvementin stealing it became obligatory that the child inquestion included that symbol. The information provided could be analysed to construct daily movements around the city at particular times and wasuseful for triangulating information gathered fromdiscussion groups.


8/13

Action research with Kampala street children 147The main disadvantage with the method was the

group nature of the activity. This was necessary inorder to hand over the design of the activity to thechildren but may have limited some children inadmitting that they took part in illegalor undesirableactivities such as stealing. Second, the omission ofdrug taking occurred in some instances because forsome street children this isomnipresent and has noparticular time or place. This, however, was animportant discovery in itself. Prior coding of theimages also meant that less detail was present in thetime lines. However, had the children drawn theirown representations, group cohesiveness and participation would have been reduced and continuityin symbols may have made later analysis difficult.With respect to the analysis of daily time lines, the

frequency of events were compared with time of dayand place of operation in order to highlight streetchild niches. In this respect the visual data provedvery fruitful. Daily time lines, however, were not souseful in eliciting oral information from children andinvolved much more probing by the researcher asthe symbols were not detailed enough to elicitinformation based on individual experiences. Therefore the children were only asked to comment onthe places where they carried out each activity. Thisallowed for the construction of route maps and theexposition of important places.

Photo diariesThe researcher's position as a white, female adultresearcher meant itwas impossible for her to enterinto the lives of Kampala street children as a fullparticipant observer. Therefore photo diaries wereused to re-create street children's' daily life processes and spatial patterns. Itwas envisaged thatimages would be produced in spaces and at timesthat the researcher would otherwise have no accessto. Fifteen disposable cameras, with flash for eveningpictures, were given to street children who wished toparticipate in the exercise. A variety of ages and bothgender groups were represented. Instructions onhow to use the camera were given to each child, orchild group. The children were asked to take picturesof the activities they did and the places they visitedover a 24-hour period. However, theywere not toldwhat or when to photograph, thereby leaving boththe content and process entirely up to the child todecide.

From the point of view of the researcher therewere four main advantages to using this method.

1. * At45

Plate 1 Photograph taken by a 12-year old boy of astreet child pick-pocketing in a busy street

Plate 2 Photograph by a 14-year old girl of street girlsbedding down for the night

First, the images produced gave excellent coverageof children's daily lives and good representation wasproduced. Second, the pictures themselves workedexceptionally well as a tool for discussion and onmany occasions the dullest and most technicallypoor pictures elicited the richest information fromthe photographer. For instance, although Plate 1represents a general street scene, insightful into howstreet children live and survive, itwas through discussion with the photographer that the main subjectwas identified. This resulted in a detailed descriptionof a street child pick-pocketing which isnot immediately obvious to the observer because the photograph is 'busy' and distant. Often such discussionswere more revealing than the pictures themselvesalthough this is not to undermine the photographstaken as some were excellent visual representationsof street life. Plate 24, for example, shows the streetspace and environment in which two street girlssleep. Third, some of the cameras went into places

where the researcher inher 'outsider' position would


9/13


L

Plate 3 Photograph taken by a 9-year old boy of hisfriend in a skip,which also captures the activities ofother street children in the background. Note the fingerof the photographer obscuring the picture

have changed the situation. Finally, often subsidiaryimages in the pictures highlighted more than the

main subjects themselves. This is demonstrated in apicture of a child standing in a skip (Plate 3). In thebackground others were sniffing fuel and collectingscrap along the roadside.

From the point of view of the child there werethree main advantages. First,most street children donot have a good self-image and initially itwas difficultfor the children to understand why theywere beingtrustedwith cameras. For example one nine-year oldboy reallywanted to participate in the research butdid not believe that he was actually being trustedwith a camera. This exercise was a major self-esteemand self-confidence builder among the children.Second, the activity gave the children access tomodern technology that would otherwise not havebeen available to them and allowed them to keeptheirown photographs. Finally, the 'fun'nature of theactivity interested many children and they becamevery excited about designing their photo diary.

As a relatively new method several issues arose.Although themajority of the cameras were returned,threewere not. In two of these cases other children

who were 'accidentally' photographed destroyedthe films. Only one camera was sold. Therefore,although there were instances where the data wasnot returned, thiswas not perceived to have invalidated the exercise. The fun aspect of this session wasenhanced by the novelty of the method, whichallowed the children to be involved in using moderntechnology, an otherwise impossibility for them.

Despite this some ethical considerations arose andon a couple of occasions the cameras caused problems between children.Where one child in a group

had been given a camera the others acted in ahostile manner towards him due to jealousy, and onanother occasion when the camera was given to agroup to share, one child ran off and kept the camerato himself, causing problems with the rest of thegroup. This only happened on two occasions andgenerally between younger children.

With regard to the actual process of taking thepictures, many of the children were not able tospread this over a whole 24-hour period. Theyounger children in particular were too excited andtended to use up all the photos in the space of a

morning or an afternoon. Although this changed thenature of the data, itwas not any less rich and in factprovided even more depth to particular activities. Inorder to avoid repetition of scenes, cameras werehanded out to different children at different times ofthe day and on different days of the week. Thein-depth nature of the photos and the subsequentdiscussion of their contents suggest that they providea valuable exclusively child-centred and child-ledsource of information.

Camera ownership also highlighted issues of concern, particularly among the younger children. Muchworry was expressed over policemen and othersecurity personnel who might think the child hadstolen the camera, and take it from them. Therefore,if required that a card containing the researcher'sname and contact details as well as a brief description of the research activity the child was taking partin, accompanied the camera. A further considerationthat arose was the use of the camera at night. Mostof the children were worried about the security ofthe camera while they were sleeping. Itwas difficultto get around this problem as itwas not possible forthe researcher tomeet with the children very late atnight when theywere about to sleep, as this iswhen

most of the children were high on drugs or fuel.The solution came from the children themselves.Cameras were given to friendly adults or securityguards, who were nearby, to keep over night. Onechild even slept at the house of a friend to avoidlosing his camera.

Visually, photo diaries produced pictures containing a wide range of situations and images. Many ofthe photographs takenwere self-explanatory both interms of the subject and setting, although somechildren were less technically adept with a camera asPlate 3 shows. In addition these diaries proved tobe a particularly useful method for stimulatingdiscussion with children. On meeting with the children after the pictures had been developed, they


10/13

Action research with Kampala street children 149were eager to talk about them and to provide verydetailed information on why they had taken eachpicture. Every picture tells a story and in someinstances this information revealed more about thesurvival mechanisms developed by street children.For example, a child who had 'snapped' a person

working inOwino Market, not only took the pictureto represent their place of work, but also because thechild in question was able to sell the photographafterwards. This demonstrates that street survivalis constantly in the forefront of the child's mind.

Such an insight would not have emerged from aninterview, which is not action-based and thereforedoes not detail the minor aspects of survival andexistence. Initially the photos were analysed inchronological sequence, to present a pictorial representation of the daily lives of individual streetchildren. These sequences were then used toconstruct a more general representation of streetchildren's' socio-spatial daily life inKampala.

The catalytic properties of visualmethodsOverall, visual methods were an importantmeans oftruly representing the socio-spatial aspects of streetlife for children in Kampala. In terms of children'soral analysis, photo diaries were particularly usefulin eliciting information. Much more detail wasextracted from 'real' as opposed to 'constructed'images such as drawings and maps. It is envisagedthat this is related to the truer representation that acolour photograph provides. Furthermore these

were reconstructions of real situations that hadoccurred in the last 24-hour period and thereforetriggered short-term memory recall. The drawingsand maps however, were mental constructions takenfrom long-term memory which has lower detailcapacity. In this instance then, photo diaries were amuch superior method for eliciting oral informationthan other pictorial representations, although theytoo elicited awealth of descriptive data.

Pictures and maps also provided good descriptiveresults owing to a higher level of detail in pictorial, asopposed to symbolic, images. This was shownthrough the map-making activities. When childrendrew pictures to represent the places they hadfrequented, ingeneral, their explanations were much

more detailed than when circles or other symbolswere used to delineate places visited. The detail wasimportant in prompting recall based on why and

what they had drawn. The daily time line activitywas

the poorest elicitor of oral description. In thisexercise the children selected to use symbolsrather than images thereby reducing the level of

mental stimulation. The group construction of thesymbols used, also reduced the level of personaldetail and thereby their usefulness as a prompt fordiscussion.

It is important to note, however, that the visualmethods described above were used in differentways and for different purposes. Eliciting oral information from children was not always a major objective of the method used. However, when a numberof visual methods are employed, they can facilitatethe triangulation of data and have proved tobe highly effective child-centred methods whenresearching the socio-spatial lives of street children in

Kampala.

Researching street children: 'real' andethical issuesHaving discussed the use of visual methods withstreet children inKampala, this section highlights twoof the main considerations thatmust be confronted

when conducting such research. These involve theproblems associated with 'real' research, inparticularthe challenge of involving street children in researchprojects and ethical research based on the moraljudgements, which often must be made.

'Real'researchFour factors arose during this research that need tobe considered in the successful collection of reliablevisual data. First,many children use drugs particularlythe inhalation of fuel. This impairs their concentrationand can result in a deterioration of the informationgiven. However, it is not always possible to workwith children who are sober, especially as themajority of full-time street children in Kampalaengage in some form of drug usage. Walking andchatting on the way to the research site, which

meant the majority sobered up significantly toengage in the research, generally counteractedthis.

Second, as street children are sporadicallyengaged in a number of daily survival strategies theresearch had to be timed and sited conveniently forthem. In Kampala, the majority of full-time streetchildren have particularworking patterns. They know

when they can get work in the markets, when theycan sweep shop fronts and generally only work long


11/13

150 Young and Barrettenough to earn enough to survive for the day. This, ittranspired,was not a large proportion of their time.

Discussions with the children prior to each activityresulted in negotiations whereby itwas possible to

meet them when they were not thinking aboutemployment opportunities and at a time and placethat did not detract from their daily survival andsocial strategies. Flexibility was thus essential, asevents and unforeseen opportunities would ariseand result in rescheduling of meetings. Therefore, it

was made apparent very early on in the research thatthe children needed to be involved in the processfrom the initial stages. This was important to ensurethe participatory and child-centred features of themethodology.

Third, lying is part of a street child's culture and isnecessary for their survival. In an interview withJames Wangobo (personal communication 1999)Drop-in Centre Manager for Friends of ChildrenAssociation (FOCA)5 itwas made clear that within afew days of arrivalon the streets, a newcomer learnsthe responses to give when asked personal questions. This iswhy it is essential to spend timewith thechildren to develop a trusting relationship and use

methods which they design and lead themselves.The visual methods, which are the focus of thispaper, fulfil these requirements well.

Finally, the research setting requires consideration.In most other research, identifying a place whereactivities can take place is not an obstacle. The streetchild, however, does not 'own' any space within theurban environment making the identification ofresearch space difficult. Initially the FOCA Head

Office was used, as there was a large tent withworktables, which was suitable. Although providing agood environment for the research some childrendid not want to go there for a variety of reasons,including distance and personal preference. Thisemphasized the importance of negotiating otherresearch locations with the children where disturbances would be minimal. Much time was spentdeliberating, discussing and testing out locationsuntil a suitable compromise was reached. This process highlighted the importance of working within afamiliar space thatwas agreeable to all the childreninvolved and illustrated that participation must beginat thismost basic level for successful and meaningfulresults to be obtained.Ethical researchEthics are an important consideration inany researchsetting but when the researcher is an outsider and

the participants an excluded population, specificissues need to be considered. Four issues central toresearching street children can be identified. First,gaining consent is vital, particularly when dealingwith children. It is important that the details of theresearch are explained at a level understandable tothe child so that they can make an informed decisionto take part (Boyden and Ennew 1997; Ennew1994). This was carefully considered when introducing visual activities and each time an activitywas organized, participation was sought from thechildren and the nature of the research explained.Issues surrounding adult consent were also pertinentto this research. For example, consent was needed

when involving female street children given thestrong protective links they have with homelessadults.A further issue was researcher positionality giventhat the researcher is not able to become fullyintegrated with the street children under investigation. In part the methods examined in this paper

were used as a way to overcome such problemsin the researcher/street child relationship. Photodiaries were used to capture scenes from aninside perspective. Drawings and maps were alsouseful in changing the nature of the research andminimizing 'outsider' involvement. Visual activitiesare particularly useful as often the children engagedwith the materials and tasks rather than with theresearcher.

Third, researcher access to this group of childrenwas important. This was facilitated by collaborationwith FOCA and further enhanced by the employment of an inside informer, a young man who hadpreviously lived on the streets of Kampala andwas well-known and respected by current streetchildren. These both provided an entry to the streetchild population, gave the researcher credibilityand were invaluable sources of information andtriangulation.

The last issue is that of data ownership. Participatory methodologies assume that data will beowned by the populations involved in the research.In this instance the children did not want to keeptheirwork and itwas not until conducting the photodiaries that the reasons for this became obvious. Themajority of street children have nowhere to keeppersonal possessions. As the photographs were anovelty many children did want to keep them, butthey could only successfully do this by asking olderfriends to look after them for them. For these reasonsthe visual data were often refused.


12/13

Action research with Kampala street children 151

ConcludingcommentsThis review of the use of four adapted visualmethods has highlighted their successful use forresearch with Kampala street children. The participatory implementation of these action-orientated

methods in non-threatening situations resulted in ahigh level of participation by children. This demonstrated the importance of placing children within theresearch process. First, children are increasinglybeing recognized as 'meaning-producing' membersof society. The use of visual methods enables theresearcher to investigate the diversity of child experience and to determine different child geographies,often producing results that adults find unexpected.Second, the fact that researchers now recognize thatthere are 'multiple childhoods', means that research

methods need to be used which can draw out suchmultiplicity. Visual methods have been demonstratedto be a very good way of including children of allages and both genders into the research process

without discriminating between those with differentabilities, confidence levels and educational attain

ments. Finally, once methods have been selected asappropriate their usefulness in eliciting the desiredinformation must be considered. As this paper hasdemonstrated, visual methods vary in the type andquality of data they produce. Photographic diaries,followed by drawings, were generally more usefulvisually and as oral catalysts, although maps and dailytime lines produce a wealth of spatial data particularly important for examining socio-spatial experiences. This equated well with the level of detail andpersonal involvement of the children in image production and allowed the images to become oralcatalysts. For example, the photographs taken by thechildren were a good method for eliciting oral detail

without subjecting the children to the authoritarianapproach that previous interview techniqueshave been associated with. This provided theresearcher with valuable data, thatwould not havebeen possible by visual or oral methods usedindependently.

These visual methods successfully allowed children to take control of the actual research processwithout the inhibitory factor of researcher presence.This is particularly important when the researcher isan outsider to the population under study. Childrenview the world differently to adults and street children have their own unique perspectives. Therefore,being able to decide, on their own initiative, about

what is important for inclusion as a representation of

street life often produces information that wouldhave been overlooked by an adult. For these reasonsvisual methods are a vital part of the researchprocess, they are fun, 'action', child-led activities thatwhen used in conjunction with other methods canincrease understanding of the socio-spatial structureof street living inKampala.Notes1 The informationwas gathered in the 1998/1999 Friends

of Children Association's (FOCA) street child census,with which the researcherwas involved.2 Under Ugandan law a child is any person under the age

of 18 years (The Children Statute of Uganda 1997).Although this is an arbitrary lassification as many children are unsure of their birthdate and thereforeguesstheirage, it is important egally.Being considered 'achild'under the law, impactson children's interactionwith thepolice and law enforcement agencies and the legalsanctions imposed.

3 'Depot' is the colloquial termused by street children inKampala to refer to the place where they sleep. Eachchild 'lives'at a specificdepot.Which effectively replacesthe conventional 'home'.

4 The eyes of the children in the photographs have beenobscured to protect their identity.

5 FOCA is a well-established non-governmental organization that has been working with street children inUganda since the early 1 80s.

ReferencesAries P 1962 Centuries of childhood Penguin Books,LondonBaker R,Panter-BrickC and Todd A 1996Methods used in

researchwith street children in Nepal, Childhood: AGloba/ Journalof ChildResearch, ChildrenOut of Place:special issue on working and street children 3 171-94

Beazley H 1997 'Streetchildren's conception and use ofspace: subcultures of resistance in Indonesia' paperpresented at the Urban Childhood Conference,Trondheim,Norway 9-12 June 1997

Bemak F1996 Street researchers:a new paradigm redefining future research with street children, Childhood: AGlobal Journal of Child Research, Children Out of Place:special issue on working and street children 3 147-56

Blades M, Blaut JM, Davizeh Z, Eiguea S, Sowen S, SoniD, Spencer C, Stea D, Surajpaul R and Uttai D 1998 Across-cultural tudyof young children'smapping abilitiesTransactionsof the Instituteof BritishGeographers 23269-77

Blaut J1997 Themapping abilitiesof young childrenAnnalsof the Association of American Geographers 87 152-58Boyden 1 1991 Childrenof the cities Zed Books, London

Boyden J and Ennew J eds 1997 Children in focus-amanual for participatory researchwith children RaddaBarnen,Swedish Save theChildren,Stockholm


13/13

152 Young and BarrettEnnew J 1994 Street and working children: a guide to plan

ning developmentmanual 4 Save theChildren,London1996 Difficult circumstances: some reflectionson 'street

children' inAfricaAfrica Insight26 203-10Golledge R G, Gale N, Pellegrino IW and Doherty S 1992

Spatialknowledge acquisition by children:route learningand relational distances Annals of the Association ofAmerican Geographers 82 223-44

HarperC andMarcus R2000 Childpoverty insub-SaharanAfrica report for Save the Children (UK)http://www.ids.ac.uk/ids/socpol/spconf/ssa.html accessed 04.02.00

Holloway S and Valentine G eds 2000 Children'sgeographies Routledge, London

IDS 1996 PRAwith children PRA topic packs 1996 Instituteof Development Studies, University of Sussex, Brighton

James A, JenksC and ProutA 1998 TheorisingchildhoodPolity Press, Cambridge

James A and Prout A eds 1990 Constructingand reconstructingchildhood FalmerPress, LondonJames S 1990 Is there a 'place' for children in geography?

Area 22 278-83- 1991 Children and geography: a reply to David SibleyArea 23 270-71

Jenks C 1996 Childhood Routledge, LondonJohnsonV,Hill Jand Ivan-SmithE1994 Listening to smallervoices: children in an environment of change ACTIONAID, Somerset

Matthews H 1984 Environmental ognition of young children: images of journey to school and home area Transactions of the Institute f BritishGeographers 9 89-1061987 Sex differences in spatial competence: the abilityof young children to map 'primed' unfamiliarenviron

ments EducationalPsychology 7 77-90

1992 Making sense of place: children'sunderstandingoflarge-scale environments HarvesterWheatsheaf, Hemel

HempsteadMatthews H and Limb M 1999 Defining an agenda for the

geography of children: review and prospect Progress inHuman Geography 23 61-90Matthews H, LimbM and TaylorM 1999 Young people's

participationand representationinsociety Geoforum 30135-44

Rudestam K and Newton R 1992 Surviving your dissertation: a comprehensive guide to content and processSage Publications,Newbury ParkCA

Sibley D 1991 Children's geographies: some problems ofrepresentationArea 23 269-701995 Geographies of exclusion:society and difference intheWest Routledge, London

Skelton T and Valentine G 1997 Cool places: geographiesof youth culturesRoutledge, LondonSwart J 1990 Malunde: the street children of Hillbrow,Johannesburg,Witwatersrand UniversityPress

Valentine G 1 96a Angels and devils:moral landscapes ofchildhood Environment and Planning D: Society andSpace 14 336-48- 1996b Children should be seen and not heard: theproduction and transgression of adults' public spaceUrbanGeography 17 205-20

- 1997 'Oh yes Ican.' 'Oh no you can't.': children andparent'sunderstanding of kids'competence to negotiatepublic space safelyAntipode 29 65-89

Wangobo J 1999 Drop-in Centre Manager for Friendsof Children Association (FOCA), Kampala, Uganda,personal communication

10 Action Research Street Children

Documents