The ESPAD validity study in four countries in 2013

Transcription

1 The Europea School Survey Project o Alcohol ad Other Drugs The ESPAD validity study i four coutries i 2013 Björ Hibell, Sabria Moliaro, Valeria Siciliao ad Ludwig Kraus I cooperatio with Arsaell Ararsso, Olga Balakireva ad Tatijaa Djurisic

2 I Legal otice This publicatio of the Europea Moitorig Cetre for Drugs ad Drug Addictio (EMCDDA) is protected by copyright. The EMCDDA accepts o resposibility or liability for ay cosequeces arisig from the use of the data cotaied i this documet. The cotets of this publicatio do ot ecessarily reflect the official opiios of the EMCDDA s parters, ay EU Member State or ay agecy or istitutio of the Europea Uio. Europe Direct is a service to help you fid aswers to your questios about the Europea Uio. Freephoe umber (*): (*) The iformatio give is free, as are most calls (though some operators, phoe boxes or hotels may charge you). More iformatio o the Europea Uio is available o the Iteret ( Luxembourg: Publicatios Office of the Europea Uio, 2015 ISBN: doi: / Europea Moitorig Cetre for Drugs ad Drug Addictio, 2015 Reproductio is authorised provided the source is ackowledged. Praça Europa 1, Cais do Sodré, Lisboa, Portugal Tel I Fax ifo@emcdda.europa.eu I twitter.com/emcdda I facebook.com/emcdda

3 I Cotets Foreword Itroductio... 5 Goals ad data collectio... 5 Key variables ad goals... 5 Participatig coutries ad schools... 6 Preparatios Results Iterest ad possible difficulties... 8 Studets uderstadig of some alcohol-related key variables Studets uderstadig of a questio about ew psychoactive substaces Testig some ew questios How hoestly the studets aswered The opiios of survey leaders Discussio Iterest ad possible difficulties Studets uderstadig of some alcohol-related key variables Studets uderstadig of a questio about ew psychoactive substaces Testig some ew questios How hoestly the studets aswered The opiios of survey leaders The validity of the ESPAD questioaire Summary ad coclusios Refereces Appedix Appedix Appedix / 41

4 I Foreword A basic questio i most research projects is whether you really measure what you wat to measure i other words, whether it has sufficiet validity. Assessig the validity of the iformatio that has bee collected is a relevat issue i ay study, especially if data are collected i more tha oe coutry. However, it becomes eve more importat whe a project is repeated over time i order to study treds that will give isight ito chages, ad the evetual relatioship of these chages to differet factors, icludig policies. Cofirmig the validity of data is therefore importat ot oly from a research perspective, but also from the perspective of developig evidece-based policies ad itervetios. Validity is a particularly critical questio i studies based o self-reports of behaviours that may ot be socially acceptable, such as the use of psychoactive substaces (drugs, alcohol ad others) amog youg people. The first data collectio of the Europea School Survey Project o Alcohol ad Other Drugs (ESPAD) was coducted i 1995 amog 16-year-old studets i 26 coutries. Sice the, data have bee gathered every fourth year i a icreasig umber of coutries. The latest ESPAD survey was coducted i 2011 whe 39 coutries took part. Data were collected for the sixth time i the witer ad sprig of 2015, ad it is aticipated that the report will be published i summer With ESPAD data from so may coutries collected i repeated surveys, validity is a crucial questio. Are data comparable betwee coutries ad over time? This questio was discussed at a early stage i the developmet of ESPAD ad a validity study was carried out i seve coutries i The coclusio was that reliability ad validity seemed to be high i all participatig coutries, though it was slightly lower i oe or two. Eve if this coclusio had bee accurate i 1998 the situatio might well have chaged sice the, ad whe preparig for the 2015 ESPAD survey it was decided to fid out whether the survey s validity was still high 15 years later. A ew validity study was therefore coducted i the autum of 2013 i four ESPAD coutries, ad more tha studets i 52 classes i 38 schools took part. We are very grateful to all the participatig studets, teachers ad school leaders for takig part i this study ad for allowig researchers access to the schools ad classes. We also would like to to express our appreciatio for the work doe by the iterviewers ad resposible researchers i the participatig coutries, ad by the authors of this report. Lisbo, March 2015 Wolfgag Götz Björ Hibell Director, EMCDDA ESPAD Coordiator / 41

5 The ESPAD validity study i four coutries i 2013 I Itroductio As soo as a Europea School Survey Project o Alcohol ad Other Drugs (ESPAD) report has bee published preparatios start for the ext data collectio, ad therefore the preparatios for the 2015 survey started whe the report from the 2011 study was published i A importat part of the preparatios was a review of the questioaire, which was the resposibility of a questioaire group. The preparatio phase for a ew data collectio is also a time for reflectig o the ESPAD methodology ad whether there are details that ca be improved or importat aspects that eed to be clarified. A aspect that is always importat to cosider is the validity of the ESPAD questioaire. I 1998 a validity study was coducted i seve ESPAD coutries i differet parts of Europe (Hibell, 1999). Studets were first asked to aswer a questioaire that cotaied core questios from the ESPAD questioaire, ad 3 5 days later they aswered a questioaire that icluded some of the questios from the first survey, accompaied by questios of a methodological character, icludig questios about how truthfully they had aswered i the first data collectio ad how truthfully they thought that their classmates had aswered. Both surveys were coducted i the classroom. I additio to this, the survey leaders aswered some questios about possible disturbaces durig the data collectio, how iterested the studets were, how seriously they worked ad how hoestly they thought the studets had aswered. The coclusio was that the reliability ad validity seemed to be high i all participatig coutries, although slightly lower i oe or two. By 2013 the 1998 validity study was 15 years old, ad the situatio may have chaged over that time. This idicated the eed for a ew study that could go ito more detail tha the 1998 study about how truthful the aswers are about the use of differet substaces. However, validity is ot oly about how truthfully the respodets aswer, but also about how well they uderstad the questios. Whether respodets give hoest aswers becomes irrelevat if they do t uderstad the questio. A ew study would provide a opportuity to explore how well studets uderstad the more difficult questios, some of which are cosidered to be key variables. I Goals ad data collectio I Key variables ad goals This first ESPAD validity study was a attempt to assess the validity of the aswers about differet substaces, without goig ito details about idividual questios. However, some questios might be cosidered more importat tha others i the sese that they are ofte used as key idicators whe lookig at treds i substace use amog Europea school studets. Variables of this kid iclude the quatity of alcohol cosumed durig the last drikig day, the frequecy of heavy episodic drikig ad the use of caabis, which is by far the most frequetly used illegal drug i the ESPAD target group (studets who will become 16 years old durig the year of the data collectio). The questio about alcohol cosumptio durig the last drikig day starts with a sub-questio that asks studets which beverages they drak durig the last day they cosumed alcohol. Separate sub-questios the follow for each of the differet beverages. There is a risk that a studet will forget that the questio is about the last drikig day ad istead give a aswer about the last day o which each of the separate beverages was cosumed. Hece there is some ucertaity over whether studets have uderstood ad aswered the questio correctly ad, if they have t, whether the use of the itroductory sub-questio as a filter questio compesated for this i the cleaig process by oly acceptig the quatities of beverages marked i the first sub-questio. Heavy episodic drikig is measured with a questio o how ofte the respodet had cosumed five or more driks i a row durig the last 30 days. The questio explais the quatity of liquid that is classed as a drik for each kid of beverage. Oe difficulty with this questio is that a studet who has bee drikig more tha oe beverage i a drikig sessio has to uderstad the quatity of liquid that is classed as a drik for each kid of beverage they cosumed; they the have to calculate how may driks of each beverage were cosumed; ad fially they have to add the umber of driks of the differet beverages together to obtai oe measure. For may studets, this will ivolve calculatig the total umber of driks per day for several cosumptio days durig the last 30 days. A alterative way of measurig heavy episodic drikig is used i the aual Swedish school surveys. The studets are asked how ofte they have bee drikig a amout of alcohol equal to at least four cas of beer, 25 cl of spirits or a bottle of wie. Whether a questio of this kid is easier for the studets to uderstad tha the traditioal five or more driks questio has bee a matter of debate. The iclusio of this alterative questio i the validity study questioaire did ot allow us to 5 / 41

6 The ESPAD validity study i four coutries i 2013 say which questio is the best measure of heavy episodic drikig, but it was a chace to fid out which of them that is better uderstood. The outcome of such a compariso was see as the basis for a discussio about the possibility of icludig both questios i the 2015 ESPAD questioaire. There is a risk that studets cosciously or ucosciously do t give accurate, hoest aswers about all substaces. I may cases this might be related to what is socially desired/ accepted i differet cultural cotexts, ad the directio of these icorrect aswers may go both ways, i.e. there may be over-reportig (exaggeratio) as well as uderreportig. Aswers about the use of illegal drugs are likely to be less reliable tha aswers about legal substaces, sice illegal drugs are cosumed to a lesser extet tha legal substaces like alcohol ad cigarettes, ad because they are illegal. The truthfuless/validity of the aswers to the caabis questio is of particular iterest, as caabis is by far the most widely used illegal drug i the ESPAD target populatio. The ESPAD questioaire cotais a questio about the hoesty of the studet s aswers to the questio about lifetime caabis use. The aswers to this validity questio idicate, as a whole, that there is a uderreportig ad that this uderreportig/hesitatio about givig hoest aswers probably differs betwee coutries. There are idicatios that this hesitacy is relatively high i may Balka coutries (Hibell et al., 2012). Whe the 2015 ESPAD questioaire was discussed with the whole group of ESPAD researchers, ad also i the questioaire group, a umber of ew ideas for questios were suggested. The questioaire group compiled a priority list of possible ew areas ad formulated some potetial ew questios. However, as the questios were utried it would be advatageous to have them tested before they were possibly icluded i the 2015 ESPAD study, ad the validity study offered such a opportuity. The primary goals of the 2013 validity study were to fid out: how well the studets uderstad the questio about quatities cosumed durig the last drikig day; how well the itroductory filter questio ad its use compesate for possible misuderstadigs of how to aswer the questio about alcohol cosumed durig the last drikig day; how well the studets uderstad the questio about heavy episodic drikig; how well a alterative questio about heavy episodic drikig may fuctio; how hoestly the studets aswer the questios about their use of tobacco, alcohol ad drugs ad their Iteret ad gamblig habits; how hoestly the studets aswer the questioaire as a whole; how hoestly they thik their classmates aswer questios about their use of tobacco, alcohol ad drugs, their Iteret ad gamblig habits ad how hoest their resposes are to the questioaire as a whole; whether the questioaire icludes ay questios that are difficult to aswer; how well the studets uderstad potetial ew questios that might be used i the 2015 ESPAD questioaire. This study tested questios about moist suff, e-cigarettes, water pipe, experiece of problems due to others drikig, ew psychoactive substaces, Iteret use ad gamblig. I Participatig coutries ad schools It was decided that a methodological study should be coducted i five ESPAD coutries i differet parts of Europe. For various reasos it was ot meaigful to draw a radom sample ad the most pragmatic method was simply to had-pick the coutries, which should be from across Europe ad should iclude coutries with differet levels of caabis prevalece. Five coutries accepted a ivitatio to take part (the lifetime caabis use i 2011 i each coutry is give i brackets): the Czech Republic (42 %), Italy (21 %), Ukraie (11 %), Icelad (10 %) ad Moteegro (5 %). Ufortuately, it became evidet that it was ot possible to carry out this research i the Czech Republic, but by the time this was apparet it was too late to iclude aother highprevalece coutry i the study. Sice the goal of the study was ot to measure the use of differet substaces i a represetative sample of studets but to lear about the validity of the aswers ad to test some questios, the aalysis did ot eed to be carried out o atioally represetative samples of schools. It was cosidered adequate to iclude studets from a umber of schools i a geographically limited area, for example i the city i which the resposible researcher lived. However, wheever relevat, differet kids of schools were icluded. It was plaed that about 200 studets would take part i each participatig coutry. It was highly desirable that the participatig studets should be the same age as the ESPAD target populatio, which meas that they were bor i If this was difficult i practice, it was acceptable for a miority of o more tha 20 % bor oe year earlier or later, i.e. i 1996 or 1998, to be icluded. As i the regular ESPAD data collectio, o more tha two classes were supposed to take part from each school. 6 / 41

7 The ESPAD validity study i four coutries i 2013 TABLE A Methodological aspects Icelad Italy Moteegro Ukraie Average Ethical aspects Ethical review ecessary No No No No Paretal coset Passive Passive No No Natioal ethical rules followed Yes Yes Yes Yes Data collectio period (2013) 4 13 November October October 7 22 October Participatio Schools ( 1 ) 38 Classes Studets Refusals Schools Classes Studets Studets presece rate 96 % 94 % 92 % 82 % 91 % Average completio time (miutes) ( 2 ) Survey leaders i the classrooms Iterveiwers ( 3 ) Teachers Iterviewers ( 4 ) Iterviewers Notes: ( 1 ) I 5 regios i differeret parts of the coutry ( 2 ) Average of reported miimum ad maximum ( 3 ) Teachers were also preset, but ot repsosible for the data collectio ( 4 ) The iterviewer resposible i the classroom did ot do ay iterview i that class A total of 9 10 classes from 5 6 schools i Icelad, Italy ad Moteegro took part i the study (Table A). The correspodig umbers were higher i Ukraie, with 23 classes from 21 schools. Oe reaso for this was that the schools were from five regios i differet parts of the coutry. No selected school or class refused to take part i the project. I Preparatios The studet questioaire cotaied both ew ad old questios, ad it was importat that at least the ew questios were traslated ad back-traslated. The same was true for the whole iterview questioaire. A traslatio backtraslatio process was performed i all coutries ad o sigificat problems were oted. Sice either the studet questioaires or the iterview form had bee used before it was importat that they were tested. I Icelad, Italy ad Moteegro this was doe with a small umber of studets i focus groups (betwee five ad 10 i each coutry), while a traditioal classroom data collectio with more tha 400 studets was carried out i Ukraie. The tests wet well ad o coutry reported that ay importat chages were eeded. Data collectio It was ot ecessary to carry out a ethical review i ay of the coutries (Table A), partly due to the ature of the study. Two coutries used passive paretal coset, while paretal coset was ot ecessary i the two others. All four coutries followed their atioal ethical rules. Data were collected i October 2013 i Italy, Moteegro ad Ukraie, ad i the first half of November 2013 i Icelad (Table A). The umber of participatig studets was aroud i Icelad, Italy ad Moteegro ad 430 i Ukraie, which totals more tha studets (Table A). Oly five studets refused to take part i the study. O average, 91 % of the studets i participatig classes were preset at the data collectio, with a lower figure i Ukraie (82 %) tha i the other three coutries (92 96 %) (Table A). The lower figure i Ukraie is about the same as i the 2011 ESPAD data collectio (Hibell et al., 2012). The studets i participatig classes received a questioaire with a majority of the ESPAD core questios plus some ew questios that were beig cosidered for use i the 2015 ESPAD data collectio (Appedix II). The questioaires were aswered i the classrooms i the same way as i the regular ESPAD data collectios, i.e. with o commuicatio betwee the respodets ad with the studets spread out i the classroom as much as possible, as for a writte test. 7 / 41

8 The ESPAD validity study i four coutries i 2013 The average time to aswer the questioaire was 27 miutes, with a rage of miutes (Table A). This is quicker tha i regular ESPAD data collectios, ad is to be expected sice the questioaire was shorter. Whe the questioaires were aswered ad put i idividual evelopes, the studets were idividually iterviewed for about 25 miutes by a experieced iterviewer. A structured iterview was performed i order to fid out how the questios i the classroom settig were uderstood, how accurately ad hoestly they had bee aswered, ad how hoestly the studets thought that their classmates had aswered. how the studet uderstood some of the substace use questios (Q7 Q10) the iterviewer did ot read out the aswerig categories to the studet, but istead iterpreted the studet s aswers by markig a fixed aswerig category. I order to stadardise the iterview situatio as much as possible i the participatig coutries, ad also betwee iterviewers, the iterviewers were traied ad required to follow the stadardised iterview istructios. The data collectio i the classrooms, ad i the iterviews, wet well i all coutries, ad o sigificat problems or disturbaces were reported. I order to limit the time take to carry out the research to two lesso hours, it was plaed that about five iterviewers would work i parallel, which meat that each iterviewer should iterview about four studets from every class. Participatio i the group admiistrated data collectio, ad i the idividual iterview, was of course volutary. The studets were guarateed complete cofidetiality ad that o results would be reported for sigle classes or schools, which meat that data would oly be reported at the coutry level. Teachers supervised the classroom data collectio i Italy, while i the other coutries oe of the iterviewers was resposible (Table A). I Moteegro the iterviewer who supervised the classroom data collectio did ot carry out ay iterviews with studets from the class they had supervised. ID umbers were required i order to lik each studet s questioaire with his/her iterview form. For the data collectio i the classroom each studet received a idividual evelope to put his/her form ito, but ot to seal. They brought the usealed evelope to the iterview ad whe the iterview started they were asked to take out the questioaire so they could look at ad refer to it durig the iterview. This made it ecessary to orgaise the iterview room i such a way that the iterviewer could ot see ay of the aswers i the studet s questioaire. The resposible researchers i all coutries have cofirmed that this was the case ad that the studets felt comfortable durig the iterviews. After the iterview, the studets put their classroom questioaire ad the iterview form i the evelope ad sealed it. ID umbers were allocated i the data etry phase whe the evelopes were opeed. The iterview was structured ad the iterviewers were asked to follow a questioaire cotaiig ready-made questios with maily fixed aswerig categories (Appedix III). For most questios the iterviewer read out the aswerig categories to the studet. However, for the questios about I Results Sice the mai objectives of the study are of a methodological character ad relate to validity, how the studets uderstood some key questios ad the testig of possible ew questios, this report maily icludes results from the iterviews, but some data from the classroom reports that the survey leaders aswered are also icluded. I additio to this, a few fidigs from the studet questioaire will be preseted i the discussio. Most tables are gathered i Appedix I ad are umbered Some methodological ad overview tables are reported i the text ad are umbered A H. A selectio of the mai results are summarised i three figures, which also are preseted i the text. Data are reported separately for each of the four coutries. I additio to this, the result tables ad graphs iclude uweighted averages. I Iterest ad possible difficulties The first questio i the iterview was about how iterestig it was to participate i the classroom data collectio. A large majority (o average early 80 %) foud it very or fairly iterestig, while oly a few (4 %) aswered that it was uiterestig (Table 1). The proportios aswerig that it was iterestig was high i all coutries ad varied betwee 61 % i Icelad ad 89 % i Italy. The studets were asked separate questios for each habit about whether they foud it easy or difficult to aswer the questios about their use of cigarettes, alcohol, drugs, the Iteret ad gamblig. For all behaviours, a very large majority aswered that they foud it very or fairly easy to aswer the questios, with averages varyig from 89 % to 95 % (Figure 1, Tables B ad 2 5). There are o major differeces betwee coutries or betwee substaces or Iteret/gamblig. The most importat fidig seems to be that slightly more 8 / 41

9 The ESPAD validity study i four coutries i 2013 FIGURE 1 Studets who foud it very easy or fairly easy to aswer questios o differet substaces ad the Iteret/gamblig Percetages amog all studets (Tables B, 2 5) Cigarettes Alcohol Drugs Iteret/ gamblig The studets who metioed that they foud oe or more questios difficult to aswer were asked which questio(s) were difficult. A large umber of questios were metioed, eight of which were metioed by more tha 10 studets (Table 8). The questio that was metioed most ofte was the oe about the frequecy of usig drugs other tha caabis (23 respodets), followed by the questio about the umber of hours spet olie (19). For six of the eight questios that studets foud difficult, a very large majority of the studets were from oe or two coutries. It is obvious from the iterviews that oe reaso why some studets foud the questio about the use of drugs other tha caabis difficult to uderstad was related to the fact that some respodets had ever heard of some of the drugs metioed i the questioaire. TABLE B Whether studets foud it easy or difficult to uderstad the questios about differet substaces, Iteret ad gamblig: proportio of studets who aswered very easy or fairly easy Percetages (Tables 2 5) Icelad Italy Moteegro Ukraie Average Cigarettes Alcohol Drugs Iteret/ gamblig Studets uderstadig of some alcohol-related key I variables As metioed above, some variables i the ESPAD questioaire are ofte used to describe substace use habits ad ca be see as key variables. A few of these questios may be slightly difficult to uderstad, icludig the questio about the amout of alcohol cosumed o the last drikig day (questio 10) ad the frequecy of heavy episodic drikig (five or more driks) (questio 11). I additio to these two questios, a alterative heavy episodic drikig questio was tested i which the amouts were described without usig the drik cocept (questio 12). studets i Icelad tha i the other coutries foud it very easy to aswer the questios about their use of differet substaces (with averages of about 86 % ad % respectively). Studets who respoded that it was fairly or very difficult to aswer oe or more questios i the four differet sectios were asked why they foud it difficult. The aswers are summarised i Table 6. The highest figure (6 %) relates to the drugs questios. I about half of these cases the questios were said to be difficult to uderstad, ad whe asked to explai this early all (29 out of 33) studets aswered that they were ufamiliar with the terms, i.e. they did ot kow about the drug(s) metioed. I additio to askig the studets how easy it was to aswer the questios o the differet substaces ad their Iteret ad gamblig habits, they were asked a separate questio about whether there were ay other questios that were difficult to aswer. O average 14 % aswered i the affirmative, with a rage from 7 % to 21 % i the four coutries (Table 7). Whe the studets were asked whether they uderstood these questios they were asked to look at their ow questioaire ad explai to the iterviewer how they uderstood it. It was the up to the iterviewer to iterpret whether this was a correct uderstadig ad mark the relevat aswer i the iterview questioaire. The questio about the amouts of alcohol cosumed durig the last drikig day cosists of several sub-questios. The first asks which beverages they cosumed that day, ad this is followed by specific questios for each beverage. There is a risk that some studets, whe aswerig the questios about the cosumptio of the differet beverages, may have forgotte that their aswer should be about their last drikig day, ad ot about the last day whe each of these beverages were cosumed. The iterviewer was asked to mark i the iterview questioaire whether the studet uderstood the questio correctly ad, if this was ot the case, whether the studet misuderstood ad did ot always thik about the last drikig day or whether he/she misuderstood the questio i some other way. 9 / 41

10 The ESPAD validity study i four coutries i 2013 FIGURE 2 Studets who correctly uderstood the questios about last drikig day, heavy episodic drikig (two questios) ad ew psychoactive substaces Percetages amog all studets (Tables C, 9 12) Latest drikig day Heavy episodic drikig (5 or more driks) Heavy episodic drikig (alterative versio) O average, 81 % of the studets uderstood the questio correctly, with a rage from 70 % i Icelad to 91 % i Ukraie (Figure 2, Table 9). O average, 10 % misuderstood ad thought that the sub-questios about the separate beverages were askig about the last day each of them was cosumed. Nearly the same average proportio (8 %) misuderstood the questio i some other way, with higher figures i Icelad (18 %) ad Moteegro (9 %) tha i Italy ad Ukraie (about 3 %). ESPAD is usig the followig questio to measure heavy episodic drikig: Thik back agai over the LAST 30 DAYS. How may times (if ay) have you had five or more driks o oe occasio? (A drik is [INSERT NATIONALLY RELEVANT EXAMPLES].) The examples describe how a drik is defied for differet beverages usig cotaiers, glasses, etc. that are atioally relevat i order to represet a cotet of about 50 cl of beer, cider or alcopops, 15 cl of wie ad 5 cl of spirits. It is importat that studets uderstad that, if they had druk more tha oe beverage, they should iclude all driks, idepedet of beverage type. However, there is a risk that the questio is misuderstood ad that some studets thik that, i order to reach the total of five driks, they should oly iclude driks of the same beverage type i their calculatio. The questio may of course also be misuderstood i some other way, ad the iterview questioaire icluded four alteratives i total. O average, 80 % of the studets uderstood the questio correctly while 11 % misuderstood that all the differet kids of beverages should be added together, ad thought that they eeded to drik at least five driks of the same beverage i order to say yes (= aswerig with a frequecy) (Figure 2, Table 10). A total of 8 % misuderstood the questio i some 72 New psychoactive substaces other way. The proportio of studets who uderstood the questio correctly varied betwee coutries, from 64 % i Icelad to 91 % i Italy. The umber of studets who did ot uderstad that they should add together all driks, idepedet of beverage type, was higher i Icelad ad Moteegro (about 17 %) tha i Italy ad Ukraie (about 6 %). There was also a larger proportio i Icelad tha i all other coutries that misuderstood the questio i some other way (14 % ad about 4 % respectively). Sice the drik cocept is ot relevat i all ESPAD coutries, a alterative way of measurig heavy episodic drikig was tested by askig the followig questio: Thik back agai over the LAST 30 DAYS. How may times (if ay) have you bee drikig alcohol equivalet to at least [INSERT NATIONALLY RELEVANT EXAMPLES]? The amout described should equal about 9 cl pure alcohol, which i reality meas cosumptio of about the same amout as i the traditioal heavy episodic drikig questio (i.e. the five or more driks versio). The examples should be atioally adjusted ad could, as a example, be somethig like four bottles of beer or a bottle of wie (75 cl) or 25 cl of spirits. O average, more tha three-quarters of the studets (77 %) uderstood the alterative heavy episodic drikig questio correctly, with a rage from 58 % i Icelad to 87 % i Ukraie (Figure 2, Table 11). The average proportio of studets who misuderstood the questio ad did ot realise that they should add together their cosumptio of differet beverages was 14 %, with higher figures i Icelad ad Moteegro (about 20 %) tha i Italy ad Ukraie (about 9 %). Fewer tha oe i te (9 %) misuderstood the questio i some other way, ad the figure was much higher i Icelad (22 %) tha i the other three coutries (about 5 %). Studets uderstadig of a questio about ew I psychoactive substaces New sythetic psychoactive substaces are ow available i may coutries as a complemet to more traditioal drugs. These substaces have become available at differet times i differet coutries, ad the types of ew psychoactive substaces have also differed betwee coutries. As this is a fairly recet pheomeo, ESPAD data collectios have ot yet icluded questios about these ew kids of substaces. Based o questios developed at the Europea Moitorig Cetre for Drugs ad Drug Addictio (EMCDDA), four were tested i the validity study (questios i the studet questioaire). I the iterview the studets were asked how they uderstood the first of these questios, ad it was the iterviewer s resposibility to iterpret whether a studet 10 / 41

11 The ESPAD validity study i four coutries i 2013 TABLE C Whether studets uderstood or misuderstood a itroductory questio about ew psychoactive substaces Percetages (Tables 12) Icelad Italy Moteegro Ukraie Average Uderstood correctly Misuderstood uderstood the questio correctly ad, if they did t, what it was that the studet misuderstood (questio 10 i the iterview questioaire). Slightly more tha seve i te studets (72 %) uderstood the questio correctly, which meas that 28 % misuderstood it i some way (Figure 2, Tables C ad 12). Eve though the proportio that misuderstood the questio varied betwee the four coutries (with a rage of %), it is evidet that the level of misiterpretatio is fairly high i all coutries. I Testig some ew questios As has bee the case whe preparig for previous ESPAD data collectios, the possibility of addig some ew questios was discussed. A questioaire group devised some potetial ew questios. Some of them had bee used i earlier school surveys, while others were more or less ew, at least i a school survey settig, which meat they should be tested before they could be used i the 2015 ESPAD data collectio. The testig of two of the proposed questios a alterative versio of the heavy episodic drikig questio ad the itroductory questio about ew psychoactive substaces has bee reported i the two previous sectios ad will ot be repeated i this sectio. The proposed ew questios were spread out i the studet questioaire ad were asked i the same sectios as other questios about the same substace. Oe was about the use of moist suff, e-cigarettes ad water pipe (questio 7 i the studet questioaire) ad the other was about others alcohol cosumptio ad its possible harm to the respodet (questio 16). I additio, five questios were asked about Iteret use ad gamblig (questios 30 34). It was thought that some studets might fid the questio about possible harm to the studet from others alcohol cosumptio a sesitive oe. However, whe the studets i the iterview were asked to report ay questios that were difficult to aswer, oly 11 metioed this questio. As reported above, oly 31 respodets (3 %) said i the iterview that the questios about the Iteret ad gamblig were difficult to aswer (Table 5). Half of these (15) metioed that they foud the iitial questio about the umber of hours they spet olie difficult to aswer. I How hoestly the studets aswered Toward the ed of the iterview the studets were asked how hoestly they had aswered the questios about their use of differet substaces, ad how much they were o the Iteret or gambled. Nearly all of the studets declared that they had aswered the questios hoestly, ad this is the case i all coutries ad also for the differet kids of questios, with proportios varyig betwee 93 % ad 100 % (Figure 3, Tables D ad 13 16). I the very few cases whe they said they had ot aswered hoestly, it was more commo for studets to uderreport tha exaggerated the frequecies. FIGURE 3 Studets who said that they had give a hoest aswer to questios about differet substaces ad Iteret/gamblig, ad their view o whether all or early all their classmates had aswered hoestly Percetages amog all studets (Tables E F, 13 20) Cigarettes Alcohol Drugs Iteret/ gamblig The idea behid testig the questio about moist suff, e-cigarettes ad water pipe was primarily ot to test whether the studets were familiar with these substaces, but to fid out if there were ay problems with the aswerig categories, which have ever bee used before i the ESPAD questioaire. Whe the studets i the iterview were asked to report ay questios that were difficult to aswer, oly six respodets metioed this questio. Ow Classmates 11 / 41

12 The ESPAD validity study i four coutries i 2013 TABLE D Studets who said that they had give a hoest aswer to the questios about differet substaces ad Iteret/ gamblig Percetages (Table 13 16) Icelad Italy Moteegro Ukraie Average Cigarettes Alcohol Drugs Iteret/ gamblig TABLE E Studets views o whether all or early all of their classmates had give hoest aswers to the questios about differet substaces ad Iteret/gamblig Percetages (Tables 17 20) Icelad Italy Moteegro Ukraie Average Cigarettes Alcohol Drugs Iteret/ gamblig TABLE F Proportio of studets who thought their classmates had maily uderreported habits Percetages amog studets aswerig that half, most or all classmates had ot aswered hoestly (Tables 21 24) Icelad Italy Moteegro Ukraie Average Cigarettes Alcohol Drugs Iteret/ gamblig N ( 1 ) ( 1 ) The umbers i each coutry vary betwee substaces ad the rages are show i the table. The figures for each substace ad coutry are available i Tables Note that the bases for the calculatios are very small, especially i Icelad. The figures are geerally high i all coutries, with Icelad the highest at 100 %. The most obvious exceptio is that a small majority (53 %) i Italy thought that their classmates had maily exaggerated their alcohol cosumptio. Relatively high figures for exaggeratio ca also be foud i Moteegro about drug use (29 %) ad i Ukraie about Iteret use ad gamblig (27 %). The studets were also asked how hoestly they thought that their classmates had aswered the substace use, Iteret ad gamblig questios. The aswers are summarised i Figure 3 ad Table E ad reported separately for the differet sectios i Tables O average, a little over 80 % of the studets said that they thought that all or early all of their classmates had aswered the questios hoestly, ad the figure is about the same for all substaces, Iteret ad gamblig. The figures are quite high for all four sectios of questios i all coutries. However, there are some importat differeces betwee coutries. For cigarettes ad alcohol, more studets thought their classmates had aswered hoestly i Icelad ad Italy tha i Moteegro ad Ukraie (o average 92 % compared with 70 %). I additio to this, fewer Ukraiia studets thought their classmates had aswered hoestly about drug use, ad the same is true about the Iteret ad gamblig figure i Moteegro. Studets who thought that half, most or all of their classmates had ot aswered hoestly were asked whether they thought that they had exaggerated or uderreported their use. I a large majority of these cases they thought that their classmates had uderreported how ofte they used differet substaces or were o the Iteret or gambled, with averages betwee 81 % ad 89 % (Tables F ad 21 24). Note that the bases for these calculatios are small, especially i Icelad, which makes the figures ucertai. I The opiios of survey leaders As is the case i the regular ESPAD data collectios, the survey leaders were asked to fill out a classroom report with iformatio about preset ad abset studets. The report also icluded questios o the survey leaders opiio of the data collectio. Iformatio from the classroom reports is preseted at class level i Tables G ad 25. I 40 of the 52 participatig classes the survey leaders reported that there were o disturbaces durig the data collectio, 11 that oly a few studets had bee disruptive, ad oe that more tha a few studets had bee disruptive. I TABLE G The opiios of survey leaders Absolute umbers at class level (Table 25) No disturbaces i the classroom All/early all aswered hoestly Very/rather easy to aswer the questioaire Icelad Italy Moteegro Ukraie Average (77 %) (92 %) (88 %) No questios difficult to aswer Number of classes (52 %) 12 / 41

13 The ESPAD validity study i four coutries i 2013 early all classes (48 out of 52) it was reported that all or early all of the studets had worked seriously. I about the same umber of classes (46), the survey leader thought that it was very or rather easy for the studets to aswer the questioaire. About half of the survey leaders (27 out of 52) aswered that there were o questios that the studets had asked about. Nie reported that some studets foud some questios that were difficult to uderstad, seve that studets did ot uderstad how to aswer a questio ad 15 reported other kids of difficulties. There is a differece betwee coutries i the umber of survey leaders who reported that studets had asked about questios. This oly happeed i oe class i Italy, while it was more commo i the other coutries. Survey leaders reportig that oe or more questios were difficult to aswer were asked which questios the studets had metioed. The examples metioed are spread out across 29 differet questios. Six survey leaders aswered that studets asked about questios 21 ad 29 i the studet questioaire, while four said questio 23. All the other questios were metioed by three or fewer survey leaders. I Discussio This sectio follows the same structure as the results sectio. Whe some of the results from a coutry differed from the results i the others, the relevat researcher was asked via to commet o these differeces. Some of these commets are preseted i this sectio. It also icludes some data from the studet questioaire. I Iterest ad possible difficulties O average, 79 % of the studets foud it very or fairly iterestig to participate i the classroom survey about their substace use, 17 % aswered either/or ad oly 4 % that it was very or fairly uiterestig. The figures clearly idicate a high level of iterest from studets i the ESPAD target populatio i takig part i school surveys about their use of differet substaces. A large majority said i the iterviews that the questios o cigarettes, alcohol, drugs, the Iteret ad gamblig were very or fairly easy to aswer. The averages were high for all four sectios, with a rage of %, which idicates that the questioaire as a whole was ot difficult to aswer. About oe i seve studets (14 %) gave examples of questios that they foud difficult to aswer, which might be see as quite a high figure. However, oly eight questios were metioed by more tha 10 respodets ad the highest umber of respodets who metioed a particular questio was difficult was 23 (of the survey participats). Hece, eve though some studets gave examples of questios that they foud difficult to aswer, the questios metioed were located i differet places i the studet questioaire ad o sigle questio stood out. For six of the eight questios metioed by more tha 10 studets as difficult to aswer, most of these studets were from oe or two coutries, idicatig cultural differeces i which questios studets foud difficult to aswer. Studets uderstadig of some alcohol-related key I variables O average, 81 % of the studets uderstood the questio about the amout of alcohol cosumed durig the last drikig day; 10 % misuderstood it ad thought that the sub-questios about the separate beverages were about the last day each of them had bee cosumed, while 8 % misuderstood it i some other way. The proportio of studets i Icelad i the last category was 18 %, which was higher tha the other coutries, ad the figure was also quite high i Moteegro (9 %). The Iceladic researcher was asked to commet o this ad respoded that may studets i Icelad do t drik alcohol, ad thus they did ot always read the alcohol-related questios carefully eough (Ararsso, 2014). A similar commet was also made by the colleague i Moteegro (Djurisic, 2014). The proportio of studets that misuderstood ad thought they their aswer should be about the last time they had druk each of the beverages ad ot about their last drikig day as such (10 %) is i practice ot a relevat methodological problem, sice the aswers to the first sub-questio about which beverages they cosumed o the last drikig day is used as a filter questio. I other words, the cosumptio of a beverage i sub-questios (b) (e) i questio 10 that was ot metioed i the iitial sub-questio is put at zero ad is therefore ot icluded i the calculatios about the amout of alcohol cosumed durig the last drikig day. Hece, the methodologically most importat studets are those 8 % who misuderstood the questio i some other way. A way to uderstad if this really is a importat methodological problem is to examie who these studets are. Data from the studet questioaire show that 76 % of these 8 % were o-cosumers (Table H). Hece, i reality 13 / 41

14 The ESPAD validity study i four coutries i 2013 TABLE H Studets who did ot drik alcohol or had ever used ew psychoactive substaces Percetages amog all studets who did ot uderstad some questios about their alcohol cosumptio or the itroductory questio about ew psychoactive substaces No-cosumers Last drikig day (misuderstood i some other way tha that the cosumptio of differet beverages was related to the last drikig day) 76 % Heavy episodic drikig, traditioal (5 or more driks) Misuderstood that beverages should be added together 100 % Other reasos 89 % Heavy episodic drikig, possible alterative questio Misuderstood that beverages should be added together 96 % Other reasos 94 % New psychoactive substaces 96 % Of the 23 % who did ot uderstad the alterative questio about heavy episodic drikig, 95 % did ot drik alcohol, which they stated i the studet questioaire (Table H). Hece, the situatio is the same as for the traditioal heavy episodic drikig questio, i.e. that early all of the studets who misuderstood the questio were o-drikers, ad as log as they state correctly that they are o-drikers it is ot methodologically importat that they do t bother to read the questio carefully eough to uderstad it correctly. As metioed above, the alterative questio about heavy episodic drikig was put i the studet questioaire to test whether it might be easier to aswer ad uderstad tha the traditioal oe. The proportio of studets who correctly uderstood the two questios was about the same, idicatig that the possible ew versio was ot easier to uderstad. Hece, sice the traditioal versio has bee used i several ESPAD data collectios, ad sice it is a frequetly used iteratioal measure, there is o reaso to chage the heavy episodic drikig questio i the ESPAD questioaire. oly about 2 % of all studets (76 % of the 8 %) did ot uderstad the questio correctly i a way that might egatively ifluece the validity, which is quite a low figure ad caot be see as a importat methodological problem. O average, 80 % of the studets correctly uderstood the traditioal questio about heavy episodic drikig (five or more driks), while 11 % misuderstood ad thought that they eeded to drik at least five driks of the same beverage i order to say yes (= aswerig with a frequecy). I total, 8 % misuderstood the questio i some other way. All (100 %) of the 11 % of studets who did ot uderstad that driks of differet beverages should be added were o-cosumers of alcohol, while 89 % of the 8 % who misuderstood the questio i some other way were ocosumers (Table H). Hece, early all of the studets who misuderstood the traditioal heavy episodic drikig questio did ot drik alcohol. Sice they aswered i the studet questioaire that they did t drik alcohol, the fact that they did ot fully uderstad the questio is ot a methodological problem. The explaatio is most probably that, sice they did t drik alcohol, they did ot (ad did ot have to) give the questio eough attetio i order to uderstad it correctly. O average, more tha three-quarters of the studets (77 %) uderstood the alterative heavy episodic drikig questio correctly. The average proportio of studets who misuderstood the questio ad did ot comprehed that they should add the possible cosumptio of differet beverages together was 14 %, while fewer tha oe i te (9 %) misuderstood the questio i some other way. Studets uderstadig of a questio about ew I psychoactive substaces A set of four questios about ew psychoactive substaces was tested. I the iterview the studets were asked whether they uderstood the first of these questios ad it was the iterviewer s resposibility to iterpret whether a studet uderstood the questio correctly ad, if this was ot the case, what it was that the studet did ot uderstad. Slightly more tha seve i te studets (72 %) uderstood the questio correctly, while 28 % misuderstood it i some way. Nearly all of the studets who did ot uderstad the questio correctly (96 %) aswered i the studet questioaire that they had ever used ay of these substaces (Table H). Hece, the situatio is the same as for the alcohol-related questios discussed above, i.e. early all of the studets who did ot uderstad the questio were o-cosumers, which they had also stated i the questioaire. May of them had ever heard about the substaces metioed i the questio, but as log as they give the aswer that they had ot used them it is ot a methodological problem that they have ever heard of them. I Testig some ew questios I additio to the ew questios discussed above, some other possible ew questios were tested i the studet questioaire. Oe was about the use of moist suff, e-cigarettes ad water pipe ad aother was about the 14 / 41

15 The ESPAD validity study i four coutries i 2013 possible harm to the respodet from others alcohol cosumptio. I additio, five ew questios were asked about Iteret use ad gamblig. Noe of these questios seemed to create ay sigificat problems for the studets, ad there are o methodological reasos why they should ot be used i the 2015 data collectio. I How hoestly the studets aswered Nearly all of the studets i all coutries respoded that they had give hoest aswers to the questios about their use of cigarettes, alcohol ad drugs, ad their Iteret ad gamblig habits. The average figure for the four sectios of questios is 98 %, which clearly idicates that, accordig to the studets themselves, they aswered the questios i the studet questioaire hoestly. It could be argued that these very high figures about oe s ow hoesty are related to the iterview situatio ad a wish to give a socially desirable aswer. However, the figures were equally high i the 1998 ESPAD validity study whe the studets aoymously aswered a self-admiistered questioaire i a classroom settig (Hibell, 1999). Hece, it seems reasoable to assume that the studets, or at least a very large majority, really did aswer hoestly. Several iterviewers reported that they thought the studets believed the data collectio was aoymous ad trusted that it was cofidetial. The importace of this for givig hoest aswers was also apparet i the 1998 ESPAD validity study (Hibell, 1999). I that study studets from two coutries (Hugary ad Swede) were iterviewed after the secod classroom data collectio, ad studets from both coutries said that they gave hoest aswers ad that the mai reaso was that they trusted that they were aoymous. More tha 80 % of all studets said they thought that all or early all of their classmates had give hoest aswers to the questios about their use of differet substaces (cigarettes, alcohol ad drugs), the Iteret ad gamblig. However, although more tha 80 % is a high figure, it is clear that some studets believed that they themselves had aswered more hoestly tha their classmates. Eve though the figures are high i all coutries for all four sectios of questios, studets belief that their classmates would aswer hoestly is slightly lower i Moteegro (cigarettes, alcohol ad Iteret/gamblig) ad Ukraie (cigarettes, alcohol ad drugs) tha i Icelad ad Italy. The resposible researchers i Ukraie ad Moteegro were asked why they thought that this was the case. The Ukraiia researcher commeted that this is probably a reflectio of the geeral mistrust of others i Ukraiia society, ad her opiio was that the truth about the hoesty of the aswers could be foud i what the studets said about themselves (Balakireva, 2014). The colleague i Moteegro thought that to some extet this was also true for the Moteegri studets (Djurisic, 2014). Studets aswerig that half, most or all of their classmates had ot aswered hoestly were asked whether they thought that their classmates had exaggerated or uderreported their habits. I a large majority of these cases (o average %) they thought that classmates had uderreported how ofte they used differet substaces or were o the Iteret or gambled. Eve though these figures are based o the aswers of a miority of about 20 % of the respodets, ad thus should be iterpreted with cautio, they suggest that studets who do t give hoest aswers ca usually be expected to have uderreported their habits. The proportio of studets who felt their classmates had uderreported their habits was geerally high i all coutries, with 100 % i Icelad as the highest. The most obvious exceptio was that a small majority i Italy (53 %) thought that their classmates had primarily exaggerated their alcohol cosumptio. Oe of the resposible Italia researchers commeted that cotrary to other substaces, the use of alcohol is ot as stigmatised i Italy ad drikig is socially accepted, so it seems plausible that some Italia studets thought that their classmate might exaggerate whe reportig their alcohol cosumptio (Siciliao, 2015). Hece, eve if the overall coclusio is that studets who do t aswer hoestly ted to uderreport their substace use, it is possible that i some particular cases uderreportig ad exaggeratig might differ betwee substaces ad coutries. I The opiios of survey leaders A large majority of the survey leaders reported that there were o disturbaces durig the data collectio, that all or early all of the studets worked seriously ad that they thought it was very or rather easy for the studets to aswer the questioaire. This idicates that the studets felt comfortable i the classroom situatio whe aswerig the studet questioaire. However, some studets asked about particular questios whe aswerig the studet questioaire. This was reported by half of the survey leaders, but to a smaller extet i Italy where it was oly metioed i oe i te classroom reports. This might idicate a cultural differece, but may also mirror that fact that Italy was the oly coutry i which teachers were survey leaders ad as such they may be more used to 15 / 41

16 The ESPAD validity study i four coutries i 2013 questios ad commets from the studets ad may ot have take as much otice of them as the research assistats who were resposible i the other coutries. The questios the survey leaders reported that the studets asked about were spread out across 29 differet questios, ad oly three were reported by more tha three respodets. Two of them were related to drug use ad, as metioed above, this is probably to a large exted coected to the fact that some studets had ever heard of oe or more of the drugs metioed i the studet questioaire. Eve though some studets asked about particular questios that were difficult to uderstad or aswer, they were spread out i the studet questioaire ad o sigle questio appeared to be difficult for more tha a few studets. I The validity of the ESPAD questioaire At least three coditios eed to be fulfilled i order to have a high validity i surveys, which i the case of ESPAD meas to correctly measure substace use habits amog Europea school studets who will become 16 years of age durig the year of the data collectio. First, the respodets must be willig to aswer the questios; secod, they must uderstad the questios; ad third, they must be willig to aswer hoestly. I the validity study, ad also i regular ESPAD data collectios, very few studets refused to take part. Hece, studet uwilligess to participate is ot a methodological problem. A very large majority foud it iterestig to aswer the studet questioaire. Feedback from the survey leaders also supported this, sice most aswered that there were o disturbaces i the classroom ad that all or early all studets worked seriously. Oly a few studets metioed i the iterview that oe or more questios were difficult to aswer. These questios were spread out i the questioaire ad o oe questio was metioed by more tha 2 % of the respodets. I additio, the survey leaders reported that the studets asked about a variety of questios, which clearly idicates that o sigle questio stads out as difficult to aswer. It ca therefore be cocluded that the questioaire as a whole was ot difficult to aswer, ad this is also supported by the survey leaders, with 46 of 52 reportig that the studet questioaire seemed to be very or rather easy to aswer. Some possible ew questios to be used i the 2015 ESPAD data collectio were tested ad oe of them was foud to be difficult to aswer. I the iterviews it was foud that some studets did ot correctly uderstad the questios about the last drikig day ad heavy episodic drikig, which have bee used i several ESPAD data collectios. This was the case for about 20 % of the studets. However, early all of these aswered i the studet questioaire that they did ot drik alcohol. Sice they were o-cosumers it seems likely that they looked for the aswer category that they did ot drik alcohol (which was easy to fid sice it was the first aswerig category) rather tha tryig to uderstad the questios about their last drikig day ad possible heavy episodic drikig. As log as a respodet aswers that he/she does ot drik alcohol, it is ot a methodological problem if a studet does ot (try to) uderstad a questio about alcohol cosumptio. As log as respodets aswer that they do ot drik alcohol, the validity is ot egatively iflueced. Quite a few studets (28 %) did ot correctly uderstad a questio about ew psychoactive substaces. I may of these cases the studets simply had ot heard about oe or more of the substaces metioed. Eve more importat is that early all of them (99 %) aswered i the studet questioaire that they had ever used ay of these substaces, which leads to the same coclusio as metioed above with regard to drikig alcohol, i.e. that as log as a studet aswers that he/she has ot used ay of these substaces, it is ot a methodological problem if the questio is ot correctly uderstood. Nearly all of the studets (o average 98 % for the differet substaces) stated i the iterview that they had give hoest aswers i the studet questioaire. As metioed above, oe might argue that this high figure might to some extet be explaied by a wish to give a social desirable aswer. However, the correspodig figure was equally high (96 %) i the 1998 ESPAD validity study whe the studets aoymously aswered a paper-ad-pecil questioaire i a classroom settig (Hibell, 1999). Cosiderig the aoymous ature of a classroom data collectio it seems reasoable to assume that the studets gave hoest aswers to questios about their substace use habits. Eve though more tha 80 % of the studets believed that all or early all of their classmates had aswered the studet questioaire hoestly, it is obvious that some respodets trust themselves more tha their classmates. I Ukraie, ad maybe also i Moteegro, this might to some extet reflect a geeral mistrust of others, which might mea that they overestimate their classmates uwilligess to aswer truthfully. It should also be oted that studets are oly guessig about their classmates hoesty, ad these guesses may ot be correct. The studets caot really kow uless they have 16 / 41

17 The ESPAD validity study i four coutries i 2013 discussed it with their classmates, ad there was o time for this before the iterview. Due to the aoymous ature of aswerig the studet questioaire i the classrooms, it seems reasoable to assume that a very large majority of the studets aswered hoestly. However, it is certaily possible that some studets did ot give hoest aswers ad that whe this happeed most of them probably uderreported. which made it ecessary to orgaise the iterview room i such a way that the iterviewer could ot see what the studets had aswered. I the iterview the studets were asked which questios, if ay, they foud difficult to aswer, how some key variable questios were uderstood, how hoestly they had aswered the questios ad how hoestly they thought that their classmates had aswered. To coclude, there is reaso to assume that the validity is high whe usig the ESPAD questioaire i a aoymous classroom settig, sice: No problems were reported from the data collectio i the classroom sessio or durig the iterviews i ay of the coutries. all but five studets were willig to aswer the questioaire; oly a few foud ay of the questios difficult to aswer; early all of the studets who did ot uderstad questios about the last drikig day, heavy episodic drikig ad the possible use of ew psychoactive substaces were o-cosumers (which they stated i the studet questioaire); there is reaso to believe that a very large majority aswered hoestly. I Summary ad coclusios The secod ESPAD methodology study was carried out i 2013 i four coutries (Icelad, Italy, Moldova ad Ukraie), followig a first study carried out i seve coutries i The goal was to study the validity of the ESPAD questioaire ad to test some possible ew questios for the 2015 ESPAD data collectio. Sice this was a methodological study, ad there was o itetio to measure substace use habits i a geographically defied area, participatig classes were ot radomly sampled. A total of studets took part 243 i Icelad, 199 i Italy, 266 i Moteegro ad 430 i Ukraie. Nie i te studets i sampled classes were preset ad participated i the data collectio. Oly five studets refused to take part. The average time studets took to aswer the questioaire was 27 miutes. I a first stage the studets aswered a questioaire that icluded may of the ESPAD core questios ad also the possible ew questios that were beig tested. This was doe i the classrooms i the same way as for regular ESPAD data collectios. Each studet placed the aswered questioaire i a idividual evelope that he/she brought usealed to a idividual iterview directly followig the sessio i the classroom. The reaso it was usealed was that the studets eeded to cosult the questioaire durig the iterview, The mai results are as follows: About 80 % of the studets foud it very or fairly iterestig to participate i the classroom survey. A large majority of the survey leaders metioed that there were o disturbaces durig the data collectio ad that all or early all studets worked seriously. A very large majority (aroud 90 %) of the studets stated that the questios i the studet questioaire were easy to aswer. A large majority of the survey leaders reported that they thought the studets foud the questioaire very or fairly easy to aswer. Whe studets were asked to metio questios that were difficult to aswer, various questios were stated. The highest umber of studets who metioed oe particular questio was 23, which correspods to 2 % of all studets. The survey leaders reported from the classroom data collectio that the questios the studets asked about were varied, ad the highest umber for sigle questios were six (two questios) ad four (oe questio). Amog studets who metioed problems i aswerig questios about drugs, the mai reaso was that they had ot heard about the drug(s) i questio. Aroud 80 % of the studets correctly uderstood the questio about the last drikig day ad also the two questios about heavy episodic drikig. I early all cases whe a questio was ot uderstood the respodet did ot drik alcohol. Close to three i te studets (28 %) misuderstood a questio about ew psychoactive substaces. Nearly all of these (96 %) had ever used these substaces. A possible ew questio about moist suff, e-cigarettes ad water pipe was tested, as was oe about the potetial harm to the respodet from others alcohol cosumptio. Five ew questios were also asked about Iteret use ad gamblig. Noe of them were difficult for the studets to uderstad. Nearly all studets (o average 98 %) declared that they had aswered the questios i the studet questioaire hoestly. 17 / 41

18 The ESPAD validity study i four coutries i 2013 More tha 80 % of all studets thought that all or early all of their classmates had aswered the questios i the studet questioaire hoestly. Most of the studets thought that ay classmate who had ot aswered hoestly would geerally uderreport their habits. The trust i classmates givig hoest aswers was slightly lower i Moteegro ad Ukraie, which to some extet might reflect a geeral mistrust of others i these coutries. The mai coclusios are as follows: Studets geerally foud the ESPAD questioaire iterestig to aswer, a opiio that is supported by iformatio from the survey leaders. With oly a few exceptios, the studets metioed that they did ot have problems i aswerig the questios, a view that also was reported by the survey leaders. Questios that studets ad survey leaders metioed were difficult to aswer were spread out across the studet questioaire ad o sigle questio stads out. There were some cultural differeces i which questios were difficult to aswer. Nearly all of the studets who did ot correctly uderstad a questio about the last drikig day ad two questios about heavy episodic drikig did ot drik alcohol, which meas that this is ot a methodologically importat issue that would ifluece the quality of the data to ay importat extet. Sice a alterative way of askig about heavy episodic drikig was o better uderstood by the studets tha the traditioal questio (five or more driks), there is o reaso to chage the traditioal versio, which has already bee used i several ESPAD data collectios. Nearly all of the 28 % of studets who misuderstood a questio about ew psychoactive substaces had ever used these drugs, which meas that this is ot a methodologically importat issue that would ifluece the quality of the data to ay importat extet. With very few exceptios, studets had o problems i aswerig the possible ew questios that were tested. It ca be expected that to a very large extet studets give hoest aswers whe aswerig the ESPAD questioaire. Whe studets do ot aswer hoestly, they probably uderreport i most cases. The possibility that studets might also exaggerate their habits was idicated with respect to alcohol cosumptio amog Italia studets, which suggests that exaggeratio will sometimes occur, ad that uder- ad over-reportig might differ betwee substaces ad coutries. Nearly all studets said that they had aswered the questios i the studet questioaire hoestly, while just over 80 % thought their classmates had also aswered hoestly. Oe importat reaso for givig hoest aswers seems to be that the studets trusted that the study was aoymous. This idicates that there are more reasos to believe what the studets say about themselves tha their assumptios with regard to their classmates. The validity is high whe the ESPAD questioaire is used i school surveys uder coditios i which the studets trust the aoymity, as demostrated by the studets willigess to give hoest aswers, combied with a very high willigess to aswer the studet questioaire, the fact that few studets foud questios difficult to aswer, ad whe they did ot correctly uderstad the questios about the last drikig day, heavy episodic drikig ad ew psychoactive substaces they aswered that they were o-cosumers. 18 / 41

19 The ESPAD validity study i four coutries i 2013 Refereces I Ararsso, A. (2014), Persoal coversatio. I Balakireva, O. (2014), Persoal coversatio. I Djurisic, T. (2014), Persoal coversatio. I Hibell, B. (1999), Do they tell the truth? A methodological study i seve coutries about the validity i school surveys. Paper preseted at the 38 th Iteratioal Cogress o Alcohol, Drugs ad other Depedecies, Viea, Austria, August 1999, The Swedish Coucil for Iformatio o Alcohol ad Other Drugs (CAN), Stockholm, Swede. I Hibell, B., Guttormsso, U., Ahlström, S., et al. ( 2012), The 2011 ESPAD Report: substace use amog studets i 36 Europea coutries, The Swedish Coucil for Iformatio o Alcohol ad Other Drugs (CAN), Stockholm, Swede. I Siciliao, V. (2015), Persoal coversatio. 19 / 41

20 The ESPAD validity study i four coutries i 2013 I Appedix 1 TABLE 1 How iterestig or uterestig it was to participate i the study Percetages Icelad Italy Moteegro Ukraie Average Very iterestig Fairly Iterestig Neither iterestig or uiterestig Fairly uiterestig Very uiterestig TABLE 2 How easy or difficult it was to aswer the questios about tobacco Percetages Icelad Italy Moteegro Ukraie Average Very easy Fairly easy Neither easy or difficult Fairly difficult Very difficult TABLE 5 How easy or difficult it was to aswer the questios about Iteret use ad gamblig Percetages Icelad Italy Moteegro Ukraie Average Very easy Fairly easy Neither easy or difficult Fairly difficult Very difficult TABLE 6 Reasos why questios i differet sectios were difficult to aswer Absoloute umbers out of all studets Questio difficult to uderstad Aswerig categories did ot fit Other All (umber) Cigarettes % Alcohol (10 1 ) 3 % Drugs (29 2 ) 6 % Iteret/ gamblig % ( 1 ) Difficult to remember ( 2 ) Not familiar with the terms All (percet) TABLE 3 How easy or difficult it was to aswer the questios about alcohol Percetages Icelad Italy Moteegro Ukraie ( 1 ) Average Very easy Fairly easy Neither easy or difficult Fairly difficult Very difficult ( 1 ) 19 % of the Ukraia studets did ot aswer the questio TABLE 4 How easy or difficult it was to aswer the questios about drugs Percetages Icelad Italy Moteegro Ukraie Average Very easy Fairly easy Neither easy or difficult Fairly difficult Very difficult TABLE 7 Whether ay questios were difficult to aswer Percetages Icelad Italy Moteegro Ukraie Average No Yes TABLE 8 The questios studets metioed as difficult to aswer Absoloute umbers out of all studets Questio umber Cotet of the questio 23 Frequecy of drugs other tha caabis 23 ( 1 ) 30 Hours spet olie 19 ( 2 ) 36 Family well off (relative to others) Father's schoolig 15 ( 3 ) 29 Possible harm of substace use Harm due to others drikig 14 ( 4 ) 13 Cosumptio of a large amout of alcohol 13 ( 5 ) 21 Difficult to get drugs other tha caabis 11 ( 6 ) ( 1 ) 20 of the 23 are from Italy ad Moteegro ( 2 ) 16 of the 19 are from Icelad ad Italy ( 3 ) 14 of the 15 are from Icelad ( 4 ) 12 of the 14 are from Italy ad Moteegro ( 5 ) 11 of the 13 are from Italy ( 6 ) All 11 are from Moteegro Number of studets metioig this questio 20 / 41

21 The ESPAD validity study i four coutries i 2013 TABLE 9 How the questio about the last drikig day was uderstood Percetages Icelad Italy Moteegro Ukraie Average Correctly uderstood Misuderstood about which day Misuderstood i some other way TABLE 10 How the regular questio about heavy episodic drikig (5 or more driks) was uderstood Percetages Icelad Italy Moteegro Ukraie Average Correctly uderstood Misuderstood beverage mixture Misuderstood examples Misuderstood the drik cocept Misuderstood i some other way TABLE 12 How the questio about ew psychoactive substaces was uderstood Percetages Icelad Italy Moteegro Ukraie Average Correctly uderstood Misuderstood examples i first bracket Misuderstood examples i secod bracket Misuderstood what the questio was about Misuderstood i some other way TABLE 13 Whether studet had give hoest aswers to questios about cigarette habits Percetages Icelad Italy Moteegro Ukraie Average Hoest aswers ( 1 ) Exaggerated Uderrated ( 1 ) Icludig the aswerig category Do t smoke, which I also aswered. TABLE 11 How the questio about a alterative way of measurig heavy episodic drikig (9 cl of pure alcohol) was uderstood Percetages Icelad Italy Moteegro Ukraie Average Correctly uderstood Misuderstood beverage mixture Misuderstood i some other way TABLE 14 Whether studet had give hoest aswers to questios about alcohol habits Percetages Icelad Italy Moteegro Ukraie Average Hoest aswers ( 1 ) Exaggerated Uderrated ( 1 ) Icludig the aswerig category Do t use alcohol, which I also aswered. TABLE 15 Whether studet had give hoest aswers to questios about drug habits Percetages Icelad Italy Moteegro Ukraie Average Hoest aswers ( 1 ) Exaggerated Uderrated ( 1 ) Icludig the aswerig category Do t use drugs, which I also aswered. 21 / 41

22 The ESPAD validity study i four coutries i 2013 TABLE 16 Whether studet had give hoest aswers to questios about Iteret ad gamblig habits Percetages Icelad Italy Moteegro Ukraie Average Hoest aswers ( 1 ) Exaggerated Uderreported ( 1 ) Icludig the aswerig category Never use the Iteret or gamble, which I also aswered. TABLE 17 Studet s view o whether classmates had give hoest aswers about their tobacco habits Percetages Icelad Italy Moteegro Ukraie Average All or early all were hoest Most were hoest About half were hoest Most were ot hoest All or early all were ot hoest TABLE 18 Studet s view o whether their classmates had give hoest aswers about their alcohol habits Percetages Icelad Italy Moteegro Ukraie Average All or early all were hoest Most were hoest About half were hoest Most were ot hoest All or early all were ot hoest TABLE 20 Studet s view o whether their classmates had give hoest aswers about Iteret use ad gamblig habits Percetages Icelad Italy Moteegro Ukraie Average All or early all were hoest Most were hoest About half were hoest Most were ot hoest All or early all were ot hoest TABLE 21 Studet s view o whether their classmates exaggerated or uderreported tobacco habits Percetages amog studets who thought that half or less of classmates had aswered hoestly Icelad Italy Moteegro Ukraie Average Maily exaggerated Maily uderreported N ( 1 ) ( 1 ) Please observe that that bases for the calculatios are small, especially i Icelad TABLE 22 Studet s view o whether their classmates had exaggerated or uderreported alcohol habits Percetages amog studets who thought that half or less of the classmates aswered hoestly Icelad Italy Moteegro Ukraie Average Maily exaggerated Maily uderreported N ( 1 ) ( 1 ) Please observe that that bases for the calculatios are small, especially i Icelad TABLE 19 Studet s view o whether their classmates had give hoest aswers about drug habits Percetages Icelad Italy Moteegro Ukraie Average All or early all were hoest Most were hoest About half were hoest Most were ot hoest All or early all were ot hoest TABLE 23 Studet s view o whether their classmates had exaggerated or uderreported drug habits Percetages Icelad Italy Moteegro Ukraie Average Maily exaggerated Maily uderreported N ( 1 ) ( 1 ) Please observe that that bases for the calculatios are small, especially i Icelad 22 / 41

23 The ESPAD validity study i four coutries i 2013 TABLE 24 Studet s view o whether their classmates had exaggerated or uderreported Iteret use ad gamblig habits Percetages amog studets who thought that half or less of classmates had aswered hoestly Icelad Italy Moteegro Ukraie Average Maily exaggerated Maily uderreported N ( 1 ) ( 1 ) Please observe that that bases for the calculatios are small, especially i Icelad TABLE 25 The opiios of survey leaders Absolute umbers at class level Icelad Italy Moteegro Ukraie Average Reported disturbaces i the classroom No disturbaces (77 %) From a few studets (21 %) More tha a few studets (2 %) Studets workig seriously All/early all (92 %) A majority (6 %) Half or less (2 %) Easy or difficult to aswer the questioaire Very easy (8 %) Rather easy (81 %) Neither easy or difficult (12 %) Rather/very difficult (0 %) Questios that studets asked about Noe (52 %) Questio difficult to uderstad (17 %) Did ot uderstad how to aswer (13 %) Difficult for other reasos (29 %) Number of classes / 41

24 The ESPAD validity study i four coutries i 2013 I Appedix 2 The Europea School Survey Project o Alcohol ad Other Drugs Questioaire o substace use Read this first please! This questioaire is part of a iteratioal study o substace use amog Europea studets. It will be aswered by studets i five coutries i differet parts of Europe. The study is called ESPAD. This is a totally aoymous questioaire. You should ot state your ame or ay other iformatio which idetifies you. You should place your completed questioaire i the eclosed evelope but ot seal it util later. After completio you will be asked to take part i a short iterview to which you should brig your evelope. Your class has bee radomly selected to take part i this study. The survey is carried out by [ORGANISATION]. It is volutary to take part. If there is ay questio you fid objectioable for ay reaso, just leave it blak. It is importat that you aswer as thoughtfully ad frakly as possible. The results will ot be preseted by sigle classes or schools ad remember your aswers are totally aoymous. If you do ot fid a aswer that fits exactly, idicate the oe that comes closest. Please mark the appropriate aswer to each questio by makig a X i the box. If you have a questio, please raise your had ad your survey leader will assist you. Thak you i advace for your participatio! Please begi. LOGO FIELD WORK ORGANISATION Cotact ifo of the orgaisatio resposible for the field work/atioal survey. 24 / 41

25 The ESPAD validity study i four coutries i 2013 The first questios ask for some backgroud iformatio about yourself ad the kids of thigs you might do C01 What is your sex? 1 Male 2 Female C02 Whe were you bor? Year 19 C03 How ofte (if at all) do you do each of the followig? Mark oe box for each lie. A few Oce or At least Almost times a twice a oce a every Never year moth week day a) Play computer games b) Actively participate i sports, athletics or exercisig c) Read books for ejoymet (do ot cout schoolbooks) d) Go out i the eveig (to a disco, cafe, party etc) e) Other hobbies (play a istrumet, sig, draw, write) f) Go aroud with frieds to shoppig cetres, streets, parks, etc. just for fu g) Use the Iteret for leisure activities (chats, music, games, etc.) h) Play o slot machies (the kid i which you may wi moey) C04 Durig the LAST 30 DAYS o how may days have you missed oe or more lessos? Mark oe box for each lie. 7 days Noe 1 day 2 days 3 4 days 5 6 days or more a) Because of illess b) Because you skipped or cut c) For other reasos The followig questios are about cigarette smokig C05 O how may occasios (if ay) durig your lifetime have you smoked cigarettes? Number of occasios or more / 41

26 The ESPAD validity study i four coutries i 2013 C06 Whe (if ever) did you FIRST do each of the followig thigs? Mark oe box for each lie. 9 years old or years years years years years years years Never less old old old old old old or older a) Smoke your first cigarette b) Smoke cigarettes o a daily basis C07 Have you ever used moist suff (sus), e-cigarettes or water pipe? Mark oe box for each lie. Yes, i the Yes, i the Yes, but more Never last 30 days last 12 moths tha 12 moths ago a) Moist suff (sus)... b) E-cigarettes... c) Water pipe The ext questios are about alcoholic beverages icludig beer, cider, alcopops (premixed driks), wie ad spirits C08 O how may occasios (if ay) have you had ay alcoholic beverage to drik? Mark oe box for each lie. Number of occasios 40 or more a) I your lifetime b) Durig the last 12 moths c) Durig the last 30 days C09 Thik back over the LAST 30 DAYS. O how may occasios (if ay) have you had ay of the followig to drik? Mark oe box for each lie. Number of occasios 40 or more a) Beer b) Cider* c) Alcopops* d) Wie e) Spirits * Optioal 26 / 41

27 The ESPAD validity study i four coutries i 2013 The followig questios are about the last day you drak alcohol C10 Thik of the LAST DAY that you drak ay alcohol. Which of the followig beverages did you drik o that day? Mark all that apply. 1 I ever drik alcohol 1 Beer 1 Cider* 1 Alcopops* 1 Wie 1 Spirits * Optioal C10a If you drak beer that last day you drak ay alcohol, how much did you drik? 1 I ever drik beer 2 I did ot drik beer o the last day that I drak alcohol 3 <50 cl cl cl 6 >200 cl C10d If you drak wie that last day you drak ay alcohol, how much did you drik? 1 I ever drik wie 2 I did ot drik wie o the last day that I drak alcohol 3 <20 cl cl cl 6 >74 cl OC10b If you drak cider that last day you drak ay alcohol, how much did you drik? * 1 I ever drik cider 2 I did ot drik cider o the last day that I drak alcohol 3 <50 cl cl cl 6 >200 cl C10e If you drak spirits that last day you drak ay alcohol, how much did you drik? 1 I ever drik spirits 2 I did ot drik spirits o the last day that I drak alcohol 3 <8 cl cl cl 6 >24 cl * Optioal OC10c If you drak alcopops that last day you drak ay alcohol, how much did you drik? * * Optioal 1 I ever drik alcopops 2 I did ot drik alcopops o the last day that I drak alcohol 3 <50 cl cl cl 6 >200 cl C10f Not at all Please idicate o this scale from 1 to 10 how druk you would say you were that last day you drak alcohol. (If you felt o effect at all you should mark 1.) Heavily itoxicated, for example ot rememberig what happeed I ever drik alcohol 27 / 41

28 The ESPAD validity study i four coutries i 2013 The ext couple of questios are also about alcohol C11 Thik back agai over the LAST 30 DAYS. How may times (if ay) have you had five or more driks o oe occasio? (A drik is [INSERT NATIONALLY RELEVANT EXAMPLES].) 1 Noe or more times C12 O how may occasios (if ay) have you bee itoxicated from drikig alcoholic beverages, for example staggered whe walkig, ot beig able to speak properly, throwig up or ot rememberig what happeed? Mark oe box for each lie. Number of occasios 40 or more a) I your lifetime b) Durig the last 12 moths c) Durig the last 30 days C13 Thik back agai over the LAST 30 DAYS. How may times (if ay) have you bee drikig alcohol equivalet to at least [INSERT NATIONALLY RELEVANT EXAMPLES]. 1 Noe or more times C14 Whe (if ever) did you FIRST do each of the followig thigs? Mark oe box for each lie. 9 years old or years years years years years years years Never less old old old old old old or older a) Drik beer (at least oe glass) b) Drik cider (at least oe glass)* c) Drik alcopops (at least oe glass)* d) Drik wie (at least oe glass) e) Drik spirits (at least oe glass) f) Get druk o alcohol * Optioal 28 / 41

29 The ESPAD validity study i four coutries i 2013 C15 BECAUSE OF YOUR OWN ALCOHOL USE, how ofte durig the LAST 12 MONTHS have you experieced the followig? If you have t used alcohol the last 12 moths, please mark zero occasios o all questios. Mark oe box for each lie. Number of occasios 40 or more a) Physical fight b) Accidet or ijury c) Serious problems with your parets d) Serious problems with your frieds e) Performed poorly at school or work f) Victimised by robbery or theft g) Trouble with police h) Hospitalised or admitted to a emergecy room i) Egaged i sexual itercourse without a codom j) Egaged i sexual itercourse you regretted the ext day C16 Have you experieced problems durig the LAST 12 MONTHS that occurred because of someoe else s drikig? Mark oe or more boxes for each lie Yes, a Yes, a Yes some- No fried or strager body else acquaitace a) Has someoe who had bee drikig called you ames or otherwise isulted you? b) Has someoe who had bee drikig harmed you physically? c) Has someoe who had bee drikig ruied your clothes or other belogigs? d) Has someoe who had bee drikig bee resposible for a traffic accidet you were ivolved i? e) Have you bee a passeger with a driver who had had too much to drik? f) Has someoe who had bee drikig harassed or bothered you at a party or some other private settig? g) Has someoe who had bee drikig harassed or bothered you o the street or i some public place? h) Has someoe who had bee drikig made you afraid whe you ecoutered them o the street? i) Has someoe kept you awake at ight with druke oise? Traquillisers ad sedatives, like [INSERT NATIONALLY RELEVANT EXAMPLES], are sometimes prescribed by doctors to help people to calm dow, get to sleep or to relax. Pharmacies are ot supposed to sell them without a prescriptio C17 Have you ever take traquillisers or sedatives because a doctor told you to take them? 1 No, ever 2 Yes, but for less tha 3 weeks 3 Yes, for 3 weeks or more The ext questios ask about marijuaa or hashish (caabis) C18 O how may occasios (if ay) have you used marijuaa or hashish (caabis)? Mark oe box for each lie. Number of occasios 40 or more a) I your lifetime b) Durig the last 12 moths c) Durig the last 30 days / 41

30 The ESPAD validity study i four coutries i 2013 C19 Whe (if ever) did you FIRST try marijuaa or hashish (caabis)? 1 Never 2 9 years old or less 3 10 years old 4 11 years old 5 12 years old 6 13 years old 7 14 years old 8 15 years old 9 16 years or older C20 Have you ever had the possibility to try marijuaa or hashish (caabis) without tryig it? 1 No 2 Yes How may times has this happeed i your life? or more The ext questios ask about some other drugs C21 How difficult do you thik it would be for you to get each of the followig, if you wated? Mark oe box for each lie. Very Fairly Fairly Very Do t Impossible difficult difficult easy easy kow a) Amphetamies b) Traquillisers or sedatives c) Ecstasy C22 O how may occasios (if ay) have you used ihalats [INSERT NATIONALLY RELEVANT EXAMPLES] to get high? Mark oe box for each lie. Number of occasios 40 or more a) I your lifetime b) Durig the last 12 moths c) Durig the last 30 days / 41

31 The ESPAD validity study i four coutries i 2013 C23 O how may occasios i your lifetime (if ay) have you used ay of the followig drugs? Mark oe box for each lie. Number of occasios 40 or more a) Traquillisers or sedatives (without a doctor s prescriptio) b) Amphetamies c) LSD or some other hallucioges d) Crack e) Cocaie f) Relevi g) Heroi h) Magic mushrooms i) GHB j) Aabolic steroids k) Drugs by ijectio with a eedle (like heroi, cocaie, amphetamie) l) Alcohol together with pills (medicamets) i order to get high m) Optioal drug* * Optioal C24 Whe (if ever) did you FIRST do each of the followig thigs? Mark oe box for each lie. 9 years old or years years years years years years years Never less old old old old old old or older a) Try traquillisers or sedatives (without a doctor s prescriptio) b) Try amphetamies c) Try ecstasy d) Try ihalats [INSERT NATIONALLY RELEVANT EXAMPLES] i order to get high e) Try alcohol together with pills (medicamets) i order to get high C25 New New substaces that imitate the effects of illicit drugs [such as caabis or ecstasy] may ow sometimes be available. They are sometimes called [ legal highs, etho botaicals, research chemicals ] ad ca come i differet forms, for example herbal mixtures, powders, crystals or tablets. Have you ever used such substaces? 1 Yes, I have used such substaces 2 No, I ever used such substaces 3 Do t kow/not sure C26 Have you used such substaces i the LAST 12 MONTHS? 1 Yes, I have 2 No I have t 3 Do t kow/not sure 31 / 41

32 The ESPAD validity study i four coutries i 2013 C27 What was the appearace/form of the ew substace you used i the LAST 12 MONTHS? Mark oe or more boxes 1 I have ot used such substaces i the last 12 moths 2 Herbal smokig mixtures with drug-like effects 3 Powders, crystals or tablets with drug-like effects 4 Liquids with drug-like effects 5 Other: C28 Thikig about your use of ew substaces i the LAST 12 MONTHS, how did you get them? Mark oe or more boxes 1 I have ot used such substaces i the last 12 moths 2 Bee give or bought from a fried 3 Bought them from specialised shop 4 Bought them from the Iteret 5 Bought them from a drug dealer 6 Other:. The ext questio asks about various substaces C29 How much do you thik PEOPLE RISK harmig themselves (physically or i other ways), if they Mark oe box for each lie. No risk Slight Moderate Great Do t risk risk risk kow a) Smoke cigarettes occasioally b) Smoke oe or more packs of cigarettes per day c) Have oe or two driks o oe occasio early every day d) Have four or five driks o oe occasio early every day e) Have five or more driks each weeked f) Try marijuaa or hashish (caabis) oce or twice g) Smoke marijuaa or hashish (caabis) occasioally h) Smoke marijuaa or hashish (caabis) regularly i) Try ecstasy oce or twice j) Take ecstasy regularly k) Try a amphetamie (uppers, pep pills, beie, speed) oce or twice l) Take amphetamies regularly The ext questios are about the Iteret, gamig ad gamblig C30 How may hours do you sped olie o a AVERAGE WEEKDAY? Mark oe box for each lie. Noe Half a hour About 1 2 About hours or less hours hours or more a) I chat rooms, sedig messages or usig social etworks ( , MSN, Facebook, Twitter, WhatsApp, etc.) b) Readig, surfig, searchig for iformatio, etc c) Dowloadig music, videos, films, etc d) Visitig adults websites e) Playig skill game (sudoku, billiard, solitaire, etc.) f) Playig games i which you may wi moey (poker, scratch, dice, ew slot, etc.) g) Participatig i Iteret games [EXAMPLES] / 41

33 The ESPAD validity study i four coutries i 2013 C31 How ofte (if ever) did the followig happe to you whe you were olie i the LAST 12 MONTHS? (If you have t bee olie the last 12 moths, please mark Never o all questios.) Mark oe box for each lie. Never Rarely From time Quite Very to time ofte ofte a) Stayed olie loger tha you iteded b) Neglected homework because you were spedig time olie c) Bee reprimaded by your parets or frieds about how much time you spet olie d) Lost sleep due to beig olie late at ight e) Felt ervous whe you were offlie ad that feelig was relieved whe you wet back olie f) Chose to sped more time olie rather tha goig out with your frieds C32 How ofte (if ever) did you gamble moey i the LAST 12 MONTHS? 1 I have ot gambled moey durig the last 12 moths 2 Mothly or less times a moth times a week times a week 6 6 or more times a week C33 How much time (if ay) did you sped gamblig moey o a typical day i the LAST 12 MONTHS? 1 I have ot gambled moey durig the last 12 moths 2 Less tha 30 miutes 3 Betwee 30 miutes ad 1 hour 4 Betwee 1 ad 2 hours 5 Betwee 2 ad 3 hours 6 3 hours or more C34 How ofte (if ever) did you gamble moey for more tha 2 hours (o a sigle occasio) i the LAST 12 MONTHS? 1 I have ot gambled moey durig the last 12 moths 2 Never 3 Less tha mothly 4 Mothly 5 Weekly 6 Daily or almost daily The ext questios ask about your family ad parets. If mostly foster parets, step-parets or others brought you up aswer for them. For example, if you have both a stepfather ad a atural father, aswer for the oe that is the most importat i brigig you up. C35 What is the highest level of schoolig your father completed? 1 Completed primary school or less 2 Some secodary school 3 Completed secodary school 4 Some college or uiversity 5 Completed college or uiversity 6 Do't kow 7 Does ot apply 33 / 41

34 The ESPAD validity study i four coutries i 2013 C36 How well off is your family compared to other families i your coutry? 1 Very much better off 2 Much better off 3 Better off 4 About the same 5 Less well off 6 Much less well off 7 Very much less well off C37 If you had ever used marijuaa or hashish (caabis), do you thik that you would have said so i this questioaire? 1 I already said that I have used it 2 Defiitely yes 3 Probably yes 4 Probably ot 5 Defiitely ot 34 / 41

35 The ESPAD validity study i four coutries i 2013 I Appedix 3 The Europea School Survey Project o Alcohol ad Other Drugs ESPAD Validity Study 2013 Iterview questioaire Explai the followig to the studets This iterview is a importat part of our desire to lear more about how Europea studets aswer our questios about substace use. It is essetial for us to kow how easy or difficult it is for studets who are your age to aswer the questios. By takig part i this short iterview about the questioaire you have just aswered you ca help us to develop a better uderstadig of how our questioaires are aswered. The iterview is totally aoymous. I do t wat you to tell me your ame. Whe we are fiished we will put the questioaire I am usig i a evelope that I will ask you to seal. It is volutary to take part i this iterview. It is importat that you aswer as thoughtfully ad frakly as possible. The substace use questioaire ad the results of this iterview will ot be preseted by sigle classes or schools, ad remember your aswers are totally aoymous. To the iterviewer Read all questios ad aswerig categories. The oly exceptios are 7 10 o which you should read the questios but ot the aswerig categories. O these questios you should iterpret the aswers. O some questios you should ask the studet to explai the aswer. Please write the aswers o the dotted lie/ope space. If ecessary, cotiue o the back of the same page. LOGO FIELD WORK ORGANISATION Cotact ifo of the orgaisatio resposible for the field work/atioal survey. 35 / 41