Smyth (1989)

Smyth, A. (1989). Measuring the Effectiveness of an Informative Audiovisual Presentation. Thèse de doctorat, University of Houston, Department of Psychology, 74 pages.

Les mêmes diapositives sont présentées seules ou en même temps que trois types de texte mettant en évidence différents aspects des diapositives. Les posttests portent sur les diapositives.

Sujets. Deux cent cinquante-trois étudiants de la University of Houston participent à l’expérience. Leur l'âge varie entre 18 et 57 ans (âge moyen de 23:3 ans). Les étudiants sont assignés aléatoirement à l'un de quatre groupes (62 à 65 étudiants par groupe).

Traitement.Le matériel expérimental comprend 266 diapositives, qui sont des photos prises à des intervalles de 22 mètres durant une promenade simulée d’environ 6 km (voir l’encadré ci-dessus). Le matériel comprend en plus trois versions d'un texte, qui sont enregistrées sur cassette audio et présentées en même temps que les diapositives. La durée de présentation du matériel audiovisuel est de 11 minutes, quelle que soit la version.

Version " Points de repère détaillés "

A1 Skycrapers in the hazy distance
B1 A service station and small shopping center
A2 A pair of gate posts
B2 mark the entrance to the subdivision
A3 A red brick home set back among the trees
B3 Its architectural style is called " saltbox "
A4 A white colonial style with dark shutters
B4 Another colonial with
A5 white shutters
etc.

Version " Trajet détaillé "

Our tour begins on the west side of Houston at the corner of Shepherd and Inwood.

We turn around and begin travelling West on Inwood away from downtown. We are on the right side of the street, looking to our left.

This is a tour rather than the most efficient route from the starting point at Inwood and Shepherd to the finish at Timberlane and San Felipe.

This route is basically from East to West, although there is some North-South travel.

At the corner of Inwood and Kirby Drive, we turn left. We are now going South. We are still on the right side of the street.

etc.

Version " Points de repère non détaillés "

Visitors to our city are usually surprised to learn that Houston, the fourth largest city in America, has no city zoning ordinances. And yet, contrary to their catastrophic expectations, within the boundaries of each residential subdivision, there are no factories, businesses, schools, churches, or apartments. Property values with a residential neighborhood tend to be stable, since the homes are usually repaired and maintained in keeping with local standards. The homes along any typical residential street are fairly uniform in size and value. The set-back distances of the houses along any given street rarely vary more than a few feet.

Unlike any other city in America, Houston residential areas are regulated by neighborhood deed-restriction instead of by a city zoning commission.

etc.

Après le visionnement des diapositives, et selon le cas, l’écoute du texte, trois tests sont administrés.

Mesure 1. Le test de reconnaissance picturale comprend 70 diapositives (28 diapositives vues préalablement, 42 diapositives semblables mais pas vues préalablement). Il s'agit d'identifier les diapositives vues préalablement.

Mesure 2. Vingt-quatre des deux cent soixante-six diapositives sont à placer en ordre chronologique.

Mesure 3. Les mêmes 24 photos, correspondant à 24 lieux, sont à placer correctement sur une carte.

Résultats. Une analyse de variance effectuée sur les données démographiques des sujets expérimentaux montre l'absence de corrélations significatives entre ces données démographiques et les mesures dépendantes.

Mesure 1. L'effet du type de texte est significatif: Var 1.2 > Var 1.3 = Var 1.4 = Var 1.1.

Mesure 2. L'effet du type de texte est significatif: Var 1.1 = Var 1.2 < Var 1.3 = Var 1.4.

Mesure 3. L'effet du type de texte est significatif: Var 1.1 = Var 1.2 = Var 1.4 < Var 1.3.

	Mesure 1: Reconnaissance picturale	Mesure 2: Ordre chronologique	Mesure 3: Emplacement spatial (carte)
Var 1.1: Texte absent	49,08 (8,00)	-1421,49 (296,79)	-812,38 (363,90)
Var 1.2: Texte (architecture)	54,36 (4,55)	-1424,27 (302,34)	-871,75 (354,00)
Var 1.3: Texte (direction, rue, etc.)	50,85 (6,40)	-1338,95 (259,92)	-673,03 (245,25)
Var 1.4: Texte (histoire)	49,83 (8,57)	-1326,39 (196,17)	-815,91 (243,37)

Discussion. " The research design, procedure and findings of this study were simple, but the underlying goals, theories and interpretations of data were complicated. On the surface, the effects of three experimental narrations on the audiences’ recognition and recall of a visual presentation were examined. As predicted, each narration was found to facilitate the group score on the dependent measure for which the narration had been designed. Further, the narrations did not affect any scores for which the narrations were expected to be neutral (i.e. produce no change). The findings showed (a) that the narrations facilitated group performances on the target dependent measures, (b) that the narrations were selective in their effects, and (c) that a " generic " narration would not similarly facilitate group performances.

The primary goal of this study was to illustrate that one could adapt schema theory to measuring the informative-effectiveness of an audiovisual presentation. This study proposed that during the construction of an informative audiovisual presentation, some of its aspects could be evaluated " scientifically ", that is, some of the variables of the elements of the audiovisual communication could be tested experimentally before the presentation is viewed by the target audience. Such testing could provide valuable information to guide the director’s choices among the alternative ways of presenting the target information " (p. 48).