Keywords: STS, STS attitude evaluation, STS opinions questionnaire, Multiple-Response Model, student attitudes toward science and technology.

• The instruments and effectively reflect prejudices of their designers, so that the researchers’ beliefs are implicitly imposed on participants by the choices offered (Lederman, 1992). Thus, the results end up attributing to participants, beliefs which are more a result of the instrument applied, than a faithful representation of their own views.
• The hypothesis of immaculate perception, which assumes that the researcher and the participant perceive and understand the text of a questionnaire in the same way (Aikenhead and Ryan, 1992; Aikenhead, Ryan, and Desautels, 1989, Lederman and O’Malley, 1990), so that the agreement or disagreement with a statement is always a result of the same reasons imagined by the designers of the instrument.
• Standardized instruments severely limit the ability to extract meaningful conclusions and evaluate changes in attitudes, since it is difficult to establish clearly what numerical score values signify an “appropriate” or “inappropriate” attitude, especially because of the low-level validity of the content (lack of correlation between what the instrument is intended to measure and what is actually measured) or because they violate the one-dimensional construct, necessary on any instrument for validating the metric results and corresponding interpretations (Aikenhead, 1988; Shrigley and Koballa, 1992).

Since the greater part of the research had been done with instruments using paper and pencil, these criticisms provoked a certain amount of precaution with regard to quantitative research, which coincided with a change of methodological paradigm toward softer, more relaxed qualitative methodologies (interviews, open questionnaires, case studies, etc.) Although qualitative methods are better suited to revealing the participants’ thinking, they also introduce biases and can hide some key aspects of the research, in particular, the researcher’s interpretations of the records obtained (Lederman, 1992).

Aikenhead (1988) addressed this problem. He compared the validity of various instruments (Likert scales, multiple-choice questionnaires closed, empirically-developed questionnaires and interviews) and concluded that the interviews provide a greater wealth of data but are time-consuming; thus, empirically-developed questionnaires based on open questions and previous interviews are a valuable third type of technique. They combine the advantages of the closed instruments, with the wealth of the interviews, save a great deal of application time, and can be applied to large samples.

Developed along these lines were the questionnaires Views on Science, Technology and Society (VOSTS) (Aikenhead and Ryan, 1989; Aikenhead, Ryan and Fleming, 1989), which has been used with students and teachers (Aikenhead and Ryan, 1992, Fleming 1987, 1988; Zoller, Donn, Wild, and Beckett, 1991a, b); and The Teacher’s Belief about Science-Technology-Society (TBA-STS) (Rubba and Harkness, 1993; Rubba, Schoneweg and Harkness, 1996), used with teachers.

Based on a taxonomy of attitudes related to science and technology (Vazquez and Manassero, 1995) and following similar patterns, the questions used in these two instruments have been adapted to the Spanish cultural context. On that basis they built the Questionnaire of Opinions on Science, Technology and Society (COCTS) (Manassero and Vazquez, 1998; Manassero, Vazquez and Acevedo, 2001, 2003a), whose implementation has suggested new methodological guidelines for improving validity and reliability, such as:

• Exchanging a single-response model (MRU), which provides minimal information about each participant’s thinking, for another of multiple choice (MRU), which allows using all the available information on each item.
• Creating new metrics, which permits one to extract from the multiple responses all the information they contain.
• Defining a standardized global attitudinal index, with an invariable metric value that reliably synthesizes the entire set of answers given.
• Classifying all the statements of the questionnaire in three categories, through an evaluation by a panel of specialist judges. This allows the application of a measure suitable for calculating attitude.

These methodological improvements allow quantitative applications such as inferential statistics and comparison of research groups in studying STS items; as well as advancing in techniques specific to qualitative research, such as is the case of personalized diagnostics (Acevedo, Acevedo, Manassero and Vazquez, 2001; Manassero and Vazquez, 1998; Manassero, Vázquez, and Acevedo, 2001, 2003a, 2003b, 2004; Vázquez, Acevedo, and Manassero, 2000, Vazquez and Manassero, 1999, Vázquez, Manassero, and Acevedo, 2005, 2006).

The purpose of the study described in this article was to evaluate the STS beliefs and attitudes of a group of high-school students through the COCTS and the new methodological procedures described. It shows the advantages and possibilities of the questionnaire and MRM applied, as well as the ability of both to assess students’ STS beliefs and attitudes reliably and meaningfully. The methodology used was based on a quantitative perspective, but also allows for qualitative analysis.

Table I. Example of a COCTS item

The multiple-choice format allows participants to express their points of views on a wide range of aspects of each topic, which may not appear in an open response, and in consequence, the attitude formed by the evaluation of the different aspects covered is quite rich and full.

The 28 items chosen for this study included 202 sentences, which should be evaluated by students, and represent the majority of the dimensions, topics and subtopics addressed by the COCTS (see Table II).

Each statement is identified by a code composed of a five-digit number (in the middle), and two letters that precede and follow that number. The middle number represents the different STS dimensions, topics and subtopics and corresponds to the code specification in the original VOSTS Table. The letter preceding the number (A, P or R) represents the category (appropriate, plausible or naïve) assigned to the statement, as rated by a judges’ scaling method. The final letter represents the relative position of the sentence within each item, listed alphabetically (A, B, C, etc.)

3.3. Procedure

The teachers of the high school STS subject applied selected items in their classes as part of a classroom activity designed to explore students’ attitudes and beliefs about STS topics. The response method applied to the questions was adjusted to a MRM, in which all the statements that made up each item assessed the degree of agreement/disagreement on a nine-point scale (Manassero and Vazquez, 1998; Vazquez Manassero , 1999).

Each direct assessment was then converted into attitudinal indices (Table III), as classified in three categories (appropriate, plausible, and naïve) previously assigned to each statement by a panel of expert judges (Acevedo, Acevedo, Manassero, and Vazquez, 2001; Manassero, Vazquez, and Acevedo, 2001; Vazquez, Acevedo, and Manassero, 2000, Vazquez, Manassero, and Acevedo, 2005, 2006).

Table III. Multiple-response model (MRM) for a COCTS item
(Signified by the direct scores ofagreement/disagreement
with each alternative statement, assignment of points on the evaluation
scale, and calculation procedures for the attitudinal indices based on the direct scores)

The MRM applied allowed us to obtain standardized attitudinal indices (-1, +1) for each sentence, according to the category assigned. Thus, the “appropriate” sentences were valued higher as they approached 9; the naïve ones as they got closer to 1; and the plausible ones (which included aspects appropriate in part) as they approached 5 (the mid-value of the scale).

Based on the indices of the statements of an item, three new indices could be calculated for each of the three sentence categories: appropiate, plausible and naïve, as an average of the previous indices for the statements in each category.

Finally, as an overall indicator of each person’s attitude toward the topic of an item, a weighted attitudinal index could be calculated (average of the indices for all three categories), following the procedure outlined in Table III.

Table IV. Statistical parameters of the distribution of average standardized attitudinal indices (-1, +1) for the sample’s answers to the set of COCTS items

4.2. General parameters of distribution

To get a better picture of student’ attitudes it was necessary to show additional details regarding the distribution of median indices for statements and items. A first step toward this was to analyze the global parameters descriptive of the group (average indices) for each of the categories of appropriate, plausible and naïve statements.

These average indices showed notable differences between the three categories. Indeed, the indices for the appropriate statements were positive, and were the highest; those of the plausible were slightly positive; and the naïve, slightly negative. Thus, the contribution of the three categories to the global attitudinal index was different. While the appropriate statements contributed in a moderately positive manner, the naïve and plausible sentences did so to a lesser extent, although in the opposite direction. On the other hand, the participating students seemed to find it much easier to identify the plausible sentences than the plausible, and even easier still than the naïve.

“In contrast, the amplitudinal range and standard deviation of the scores for each category, showed an order opposite to that above, with the naïve statements ranked highest, and the appropriate ones, lowest. This pattern of overall variation between categories suggests that students’ adherence to the more appropriate positions of the STS items seemed to be a little easier than that of naïve and plausible sentences.

The above results demonstrate how to what point the attitude measurement could change significantly depending on the category of the statement applied. This issue has important methodological implications for the validity of attitude-evaluation instruments, and will be addressed in the discussion section.

4.3. Strengths and weaknesses of STS attitudes

It is possible to perform qualitative studies of specific questions by analyzing the mean indices of each sentence of an item. These provide direct information about the overall attitude of the sample (or of each person) regarding the belief expressed in each sentence. The highest and lowest attitudinal scores for sentences are indicators of students’ strongest and weakest attitudes, respectively.

High scores indicate beliefs consistent with a proper understanding of science and technology, from a current perspective of the philosophy, history and sociology of science, while low scores represent the most dissonant ideas (see Table V).

Sentences with the highest scores represent the beliefs of those students who are more in accord with an appropriate understanding of science and technology. The five sentences with the highest positive scores, that meet stringent criteria for superseding the sample mean by more than two standard deviations are mostly naïve (shaded in Table V). Among them are the following four beliefs:

• Science cannot be defined (I10111I).
• Science does not have much influence on technology (I10412A).
• Technology is does not have very much influence on society (I40811A).
• Science does not have very much influence on society (I40821A).

All these are easily-accepted claims in a developed society, where science and technology are constantly present in everyday life and in the society (e.g., in social communications media and in advertising), and therefore do not need an excessive amount of communications and technological knowledge for their recognition. The appropriate statement (A50111E) which states that “there are not just two types of people (scientists and non-scientists), since as there can be as many kinds of people as there are individual preferences, including those who understand both science and letters;” since this also seems to be a fairly common idea in contemporary society, the high value of the index is not too surprising.

If there is applied a less demanding criterion that exceeds the highest scores from the sample mean on more than one standard deviation, 29 sentences are obtained (Table
V). Among them there are 15 appropriate, 11 naïve and only 3 plausible statements, so that most of the participating students’ best beliefs correspond to statements classified as appropriate by the judges. This analysis reconfirms the idea that different statements are not equivalent for the diagnosis of attitudes: the appropriate and naïve statements are much better than the plausible ones identified by students.

At the other extreme, the sentences with the lowest scores represent students’ beliefs less in accord with a good understanding of science and technology. They could be interpreted as alternative ideas about the nature of science, with respect to current knowledge about philosophy, history and the sociology of science.

In this case, one can also use two identification criteria. If there is applied the less stringent criterion that scores are lower in more than one standard deviation from the sample’s mean, 40 statements are obtained. These appear in Table VI. Of these, 28 are naïve sentences, 11 are plausible and only one is appropriate.

Only one sentence meets the most demanding criterion: that the score be lower than the sample mean in more than two standard deviations. This sentence corresponds to the familiar naïve belief that technology is applied science, which appears formulated in a slightly different manner in three other statements with significant negative indices.

Among other alternatives there are several other naïve beliefs about the relationship between science, technology and society, such as that scientific research continues regardless of cultural or ethical views, and regardless of the clashes between scientists and religious or cultural groups, or on the other hand, that the social demand for understanding nature stimulates the accumulation of scientific knowledge.

A tendency toward scientism appears when it is admitted in a naïve manner that science consists in “discovering new ideas” or “inventing things for the benefit of society.” There is also an over-reliance on scientists when one considers that most of them are worried about the possible adverse effects of their discoveries, have the training and information that give them better understanding of the issues, or have the knowledge and do not act out of personal interest.

Other outstanding naïve beliefs are that science is available for the use and benefit of everyone; but there should not be more science studied in school, because that would not give students the ability to understand it better or improve their interest in it. Furthermore, it is believed that one cannot force people to study science because some people do not like it; that not everyone can understand it; and that others should not decide for students what they have to study.

The most naïve ideas concerning the epistemology of science, which can be seen in Table VI, also show the credulity of students in believing in the absolute and cumulative character of scientific knowledge, a certain unsophisticated realism imbued with more conviction regarding discovery than concerning the invention and construction of scientific knowledge, as well as belief in the ability of experiments to test hypotheses, in theories as mature laws and in the existence of a universal scientific method.

All in all, students in this study show a collection of naïve STS beliefs, which include some that are consistent with the results of previous research on the nature of science, and others that provide new profiles of students’ STS.

4.4. Diagnosis of attitudes relating to an STS item

Based on the standardized attitude indices obtained for each statement, there can be calculated for each item a global index, which represents the attitude toward the issue raised therein. This calculation can be made by using two different procedures. The first calculates the arithmetic mean of the attitudinal indices of all the statements making up each item (overall mean attitude index.) The second, a little more sophisticated, consists in first calculating the averages for the set of statements by categories (appropriate, plausible and naïve) of the item separately, so as to obtain three indices (one for each category). Then one calculates the average of these (weighted global attitudinal index), following the procedure described on the right side of Table III. For various reasons, mostly due to better weighting of the various categories, this second method is the one applied despite its greater complexity (Manassero, Vázquez, and Acevedo, 2004; Vázquez, Manassero, and Acevedo, 2005).

Similarly, one can repeat the qualitative analysis for each individual sentence in order to identify the most powerful aspect and the weakest, and for each of the categories, by analyzing the mean indices of these, for the various items. As an example, in Table VII there is a summary of the calculation of the indices for three items, done using both methods: one with a large number of plausible sentences (10211), another with a large number of naïve sentences (90,611) and a third more balanced as to the types of sentences (91011). The indices of the three categories assume an approach to the overall thinking of the sample (or that of each student); this highlights the strengths and weaknesses relative to the more naïve or more appropriate beliefs about a question.

Table VII. Index values for the three Items

The results of the global index for the items listed are similar (around zero), but when the scores of the three categories are analyzed, the following aspects stand out:

• The first item (10211) repeats the general pattern of the overall sample, with a moderately low index for the appropriate sentences, an index of close to zero for the plausible ones, and a negative index for naïve sentences. The strongest aspect of this item is the recognition of appropriate ideas, while the weakest the inadequate identification of naïve statements.
• The index for the scientific method (90611) has its strength in identifying plausible sentences, but fails to recognize appropriate and naïve ideas.
• By contrast, the attitude toward the nature of discovery or invention of scientific theories (91011) has the strongest point in the index of plausible sentences, while the weakest is that of the naïve statements.

If this analysis based on quantitative indices is extended to the individual sentences also, new results and details of assessments of attitudes can be obtained. For example, in the category plausible of Item 10211, although the average is almost zero, there are differences between the scores of the various statements of this type, some clearly positive and some clearly negative. So this analysis—a bit more thorough—can come even closer to the achievements and shortcomings of the group (or persons considered individually.)

From a teaching perspective, there is an obvious educational value for this type of evaluation and a qualitative analysis based on the quantitative data obtained. Thus, for the above items, the planning of teaching requires activities of a different sort depending on the strengths and weaknesses. In this case, the most common weaknesses relate to the naïve beliefs; in others, it might require a methodology of conflict between good and naïve beliefs.

4.5. Diagnosis of attitudes toward an issue

In addition to qualitative and quantitative microanalyses of each issue, such as those outlined above, it is also possible to conduct a larger study focused on a specific STS dimension (e.g., attitudes toward the influence of society on S & T) by analyzing a set of items belonging to the same dimension. Here are some results taken from the responses that participating students gave to the questions in each dimension (Table VIII).

a) Definitions of science and technology

The dimension Definitions of Science and Technology was evaluated by four questions on the definition of science (10111), the definition of technology (10211), and the mutual dependence between science and technology (10412, 10413). To complete it, there was also included in this dimension a question concerning the interaction between science, technology and society (30111).

All in all, attitudes toward this dimension were low, but positive. The more positive overall attitudes toward the questions (about +0.20) were obtained regarding the definition of science; the influence of technology on science; and the interaction between science, technology and society. In contrast, the lowest overall attitudes (around zero) were obtained for the definition of technology, and the influence of science on technology.

The general profile of the attitudinal indices by category showed some similarities and some differences. The similarities are given in the scores for the appropriate statements, which were more positive in all cases (indices between +0.2 and +0.4), and the scores for the plausible sentences, which were grouped at values close to zero (between 0 and +0.1). The differences appear in the indices for naïve statements, notable for their wide dispersion of positive and negative values, ranging from a very positive value in the case of the definition of science (near +0.5) to a very negative value for the definition of technology (close to -0.3).

The profiles of variation between categories declined in three items, going from the appropriate statements to the naïve, passing through the plausible (intermediate), while the profiles of the other two items showed a variation in V (appropriate: high; plausible minimum ; naïve: high). All in all, the overall profile of this dimension showed the highest rates in the appropriate statements, intermediate (near zero) for the plausible and widely dispersed in the naïve.

b) External Sociology of Science

The dimension External Sociology of Science was the most extensive in the number of subjects represented. It comprises two subdimensions: the “influence of society on science and technology” and the “Influence of science and technology on society.”

Included in the first subdimension were ethical issues (20411), educational institutions (20,511), special interest groups (20611) and the mutual influence between society and science and technology (20811, 20821). Overall attitudes varied widely, depending on the topic addressed in the questions. The question with the highest score (greater than +0.20) was that related to the general influence of society on technology (20811). Those for special interest groups (20611) and the general influence of society on science (20821) were moderately positive. The attitudinal index was practically null on the question related to studying more science and technology in educational institutions (20511), and moderately negative in the question about the influence of ethical beliefs on science (20411).

Indices by categories showed a profile similar to the previous dimension (definitions of science and technology): the more positive indices appeared in the evaluation of appropriate sentences (about +0.2), followed by plausible sentences, in this case with a greater dispersion (grouped around values close to zero, from a slightly negative value +0.1), and widely dispersed values in the indices of naïve statements, which ran from a very positive value for the naïve statements regarding the influence of society on science (+0.4), to very negative for naïve statements about the influence of ethical beliefs (about -0.3).

Three profiles of variation decreased from the appropriate statements to the naïve, passing through to the plausible (intermediate), while two others showed a variation in V (appropriate: high; plausible minimum, naïve high). In summary, this subdimension showed the highest rates in the category appropriate, close to zero in the plausible, and widely dispersed in the naïve.

In the second sub-dimension were included issues of social responsibility (40111), social decisions (40211), social problems (40311), economic welfare (40511) and the general influence of science and technology on society ( 40811, 40821).

The indices of global attitudes (mean and weighted index of the items) for these questions were very similar and moderately positive (between +0.1 and +0.2), except for the question concerning social decisions (slightly negative). The indices of the three categories showed scores minimally grouped and having great variation between the different items within each category. In the case of the appropriate statements, scores were positive for all questions except 40821 (index zero), while in the naïve and plausible, they were both positive and negative.

The variation across the three categories showed a dominant profile of decline in half the items (40111, 40211 and 40311) from the category appropriate (most positive) to the naïve (more negative or less positive) with an intermediate position for the category plausible. One item (40821) showed an inverse profile, rising from the category appropriate (least positive) to the naïve (most positive). The other two items (40511 and 40811) had a profile in V (high in the appropriate statements, minimum in the plausible, and high in the naïve).

The more positive indices appeared in the assessment of the appropriate sentences (between 0 and +0.4), followed by the plausible sentences (between +0.2 and -0.2), and were more scattered in the naïve statements, going from a clearly positive value at times (greater than +0.2), to a quite negative value on the issue of making social decisions (-0.3).

In summary, this subdimension showed the highest rates in the category appropriate, close to zero in the plausible, and widely dispersed in the naïve, but with wide variation in scores between the different items within each category.

As a further aspect of the influence of science and technology on society, there were also applied some specific questions related to the influence of in-school science on society, such as its contribution to the unification between scientific and humanistic culture (50111); the usefulness of in-school science (50211) and the scholastic description of science (50311). The indices of global attitudes (mean and weighted) for these items was very low, especially for the last two, and different from each other (positive, almost null to slightly negative.) The variation across the three categories showed profiles very irregular and different from those described above.

The indices of the items in the three categories showed great diversity. In the case of the appropriate statements, scores ranged from a very positive maximum (+0.5) and a quite negative minimum (-0.2); in the naïve all were negative (between 0 and -0.1), and the plausible ranged between positive and negative values close to zero.

In brief, the issue of in-school science’s influence on society showed rates close to zero in most cases, and an irregular variation in the scores for the different items, resulting in relatively lower attitudes on the whole.

c) Internal sociology of science

The internal sociology of science included two items relating to the characteristics of scientists, such as their motivations for doing research (60111), and the effects of gender on science (60511). It also included an item related to the social construction of scientific knowledge, having to do with the decisions made by scientists (70211), and two pertaining to the social construction of technology, which referred to technological choices (80131) and the autonomy of technology (80211).

The results for this dimension presented a clear difference between the item related to the effect of gender on science (60511) and the others, for which reason it will be treated separately. The gender item presented the only negative attitude (-0.03) of this dimension as a direct mean of the sentences, and at the same time, the most positive (+0.20) as a weighted index. Indices by categories show a V variation (high in the appropriate statements (maximum for the dimension); minimum in the plausible (negative); and high in the naïve (also the maximum of the dimension).

The rest of the items of this dimension presented global parameters similar to those observed; but with a new profile of variation among the three categories. Global attitudinal indices, both mean and weighted averages, were grouped in a moderately positive band below +0.20.

The profile of the indices under the three categories showed a similar positive score for the appropriate and plausible statements, followed by a sharp drop to a negative minimum for naïve sentences.

In summary, the dominant profile of the attitudes in this dimension (with the exception of the gender-related) showed similar indices (higher and positive) in the categories appropriate and plausible, and a negative score in the naïve categories.

d) Epistemology

The dimension Epistemology, also called (in reductionist mode) “The Nature of Science” by many authors in the bibliography, contained many essential issues concerning the nature of scientific knowledge. It included topics relating to the nature of scientific models (90211), status and relationships between hypotheses, theories and scientific laws (90511), the approach to scientific methodology (90611) and the epistemological status of scientific knowledge (91011).

The results of this dimension presented the students’ clearest insufficiencies, since the overall mean indices are the lowest, and are distributed around zero. The item dealing with scientific models has the most positive overall index of this dimension, although it is quite low (+0.10); and the one addressing the status and relationships between hypotheses, laws and theories presented the most negative overall index ( -0.10). Indices by categories show an approximately inverted V variation, somewhat asymmetric (low in the appropriate statements, highest in the plausible, and low in the naïve).

In a word, this dimension shows positive and negative indices that are quite low, around zero, higher and more positive in the category of plausible sentences, and lower and negative in the naïve statements.

The overall indices of the various Items of epistemology suggested that the attitudes were based on the specific aspect addressed in each case. Thus, while in Item 90211 (epistemology of scientific models) the score was moderately positive, in 90511 (status and relationships between hypotheses, laws and theories) it was negative. The other two items, 90611 (nature of the scientific method), and 91011 (status of scientific knowledge) had intermediate indices.

Indices by categories also showed strengths and weaknesses in each issue, which could be analyzed according to the schemas already mentioned. Items 90211 and 91011 had indices in different categories with a negative charge on the naïve statements, and a positive on the appropriate and plausible. However, issue 90611 had negative indices in the categories of appropriate and naïve statements, while it had positive ones for the plausible sentences. All in all, the methodology used to assess the overall attitude in every STS dimension allowed us to diagnose the strengths and weaknesses, as already shown in previous sections. This will be completed in later analyses

4.6. Diagnosis of individual attitudes

Quantification of the responses through the use of various attitudinal indices also allows personalized diagnoses, typical of qualitative research. The individual diagnosis can be made from two different levels:

• Relative, which consists in the evaluation of a person’s attitude by other members of the group. The general reference could be the average of the whole group (in small samples, the person being compared should be excluded to improve reliability).
• Absolute, which is the individual attitude regarding the profile determined by the category which judges assign to sentences, i.e. the attitudinal score itself. The reference here is the maximum suitability score for each sentence, represented by the optimal attitude profile (set to the value +1 for all statements and items).

Taken as a whole, the index values of the group are distributed in all the statements along the whole range of attitudinal indices from the maximum rates (+1) to the minimum (-1), covering all the topological space of scores on all the items. This plurality of beliefs emphasizes the great diversity of the group’s individual attitudes, which may indicate some confusion and lack of training in the area evaluated (Figure 1).

Figure 1. Comparison of the attitudinal profiles of the statements of Item10211 (definition of technology), for a student with a very positive attitude and another with a very negative attitude (as a reference, there is also shown the group’s attitudinal average.

For example, as related to the question raised by the definition of technology (10211), Figure 1 depicts three attitudinal profiles belonging to the total group average and to the responses of two students who have extreme attitudinal scores on this issue, i.e., a very good attitude (highest mean attitudinal index) and a very poor one (very low mean attitudinal index). Similarly, in order to facilitate understanding, it could represent any other individual attitudinal profile not included in the figure.

Logically, the profile of the person with low attitude exhibits indices with a minimum evaluation level in many sentences, with the notable exception of sentences E and G, where the individual obtained a maximum evaluation (in the last, it is even superior to the score of the other student, who had a good attitude). The profile of the person with the better attitude is less homogeneous than the previous one, because the indices of the different statements take very different values, although none reached the minimum value.

These two examples of individual evaluation show, simultaneously, the complexity and wealth of the attitudes towards STS issues, since one person may obtain opposite scores on the sentences. At the same time, Figure 1 demonstrates once more the methodological problem of the ambiguous value an attitudinal assessment can have based on a single sentence. If a student with a negative attitude were evaluated only by her response to the statement E (or G) she would get a maximum attitude, and vice versa, if a student with the best attitude were evaluated only for her response to the statement B, she would get a negative attitude.

1. Research continues in spite of religious and cultural groups, and regardless of cultural or ethical views.
2. Scientists are concerned about the possible adverse effects of their discoveries, and verify them.
3. Scientists make better techno-scientific decisions because they have training, data and knowledge.
4. The decision to use a new technology depends mainly on its benefits to society.
   
   The results obtained prove that students’ thinking can be contradictory and inconsistent, a situation nothing strange in attitude-evaluation studies. Indeed, an analysis of the different dimensions shows that various items of the same dimension can lead to very different values, and that even within the same item, the different phrases also present very different indices. One might think that a person’s having either an appropriate attitude or a naïve one concerning an item would mean that this attitude would remain more or less homogeneous in all the sentences that comprise it. On the contrary, the results show a wide dispersion, revealing inconsistency in the students’ attitudes, without there seeming to have arisen any sort of personal conflict. Consequently, it is necessary to educate STS attitudes, addressing them in an explicit, reflective manner in the science curriculum.
   
   This study shows the complexity of attitude formation, and provides additional evidence on the methodological precautions that must be taken in order to obtain a more valid and reliable attitudinal assessment (Munby, 1997, Gardner, 1996; Ramsden, 1998). The risk of invalidity inherently associated with means of evaluating attitudes is great when one attempts from a few sentences (sometimes with one sentence only), to infer an attitude toward a more general or broad subject. Indeed, if the contribution to the definition of the participants’ attitudes may be skewed, depending on the content of the sentence, it is obvious that the validity of attitudinal means based on responses to a series of sentences depends on the representativity of the statements applied, either in a traditional Likert questionnaire or in an interview. Consequently, a single sentence can skew the measurement obtained, so that the use of a single item to assess an attitude must be avoided.
   
   The discrepancy observed between the opinions of the sentences of the same item (or the various specific issues of the same dimension) clearly demonstrates the limited validity of assessment methods based on using isolated sentences to assess attitudes, and the need for valid, but necessarily complex methods, to obtain a more reliable assessment and stronger results.
   
   The method proposed in this study, which evaluates STS attitudes appropriate to specific topics based on the indicators of a diverse set of sentences related to a topic, fulfills the same functions as do the similar ones used in an interview to clear progressively the participant’s thinking, but it does so in a more structured and premeditated manner.
   
   5.1. Educational Implications
   
   The individual diagnosis of STS attitudes has important implications for science teaching. A diagnosis of this type facilitates the planning of didactic training activities aimed toward improving or changing STS attitudes, and toward adaptation. This possibility is especially valuable when combined with the plurality and diversity of beliefs observed in the students.
   
   The major differences between the STS attitudes of a relatively small set of participating students suggest that the most appropriate training action would be more effective if it were more personalized; that is, adapted to the diverse needs of each person. In this sense, the detection of strengths and weaknesses is key, since educational intervention should be supported by the strongest, and aimed at improving the weakest.
   
   Individual diagnostic accuracy combined with the plurality of attitudes obtained, provides a valuable resource for educational intervention aimed at changing attitudes. Research on attitudinal change has shown that one of the most powerful factors in achieving this is peer influence. One possibility would be to organize work and discussion groups among students, i.e. learning communities led by those who both have the right attitudes, and are the learners’ equals. Individual diagnostic assessment allows the identification of students having the most appropriate attitudes in every sentence, item or dimension. Furthermore, these students could lead others in learning, in those aspects they master more completely.
   
   Finally, to complement the didactic activity and to achieve the desired attitudinal change, these might also help: appropriate and suitably selected readings for raising the awareness of logical inconsistencies between more appropriate ideas and less appropriate ones; dealing with the conflict between opposing ideas in group discussion; and explanations designed to address each of the weaker aspects.
   
   Educational intervention could be performed using various methods and instruments belonging to the field of general didactics itself. Explicit personal reflection should be, in any case, a basic tool for achieving attitudinal change, regardless of the teaching method used.
   

Aikenhead, G. S. & Ryan, A. G. (1992). The development of a new instrument: “Views on science-technology-society” (VOSTS). Science Education, 76 (5), 477-491.

Aikenhead, G. S., Ryan, A. G., & Désautels, J. (1989). Monitoring students’ views on science-technology-society topics. Work presented at the Annual Meeting of the National Association for Research in Science Teaching, San Francisco, CA.

Aikenhead, G. S., Ryan, A. G., & Fleming, R. W. (1989). Views on science-technology-society (form cdn.mc.5). Saskatoon, Canada: University of Saskatchewan, Department of Curriculum Studies. Retrieved January 10, 2006, from: http://www.usask.ca/education/people/aikenhead/vosts.pdf

Ben-Chaim, D. & Zoller, U. (1991). The STS outlook profiles of Israeli High-School students and their teachers. International Journal of Science Education, 13 (4), 447-458.

Eagly, A. H. & Chaiken, S. (1993). The psychology of attitudes. Forth Worth, TX: Harcourt Brace College Publishers.

Fleming, R. W. (1987). High-school graduates’ beliefs about science-technology-society. II. The interaction among science, technology, and society. Science Education, 71 (2), 163-186.

Fleming, R. W. (1988). Undergraduate science teachers’ views on the relationship between science, technology and society. International Journal of Science Education, 10 (4), 449-463.

Gardner, P. L. (1996). The dimensionality of attitude scales: a widely misunderstood idea. International Journal of Science Education, 18, 913-919.

González-García, M. I., López-Cerezo, J. A., & Luján, J. L. (1996). Ciencia, Tecnología y Sociedad. Una introducción al estudio social de la ciencia y la tecnología. Madrid: Tecnos.

Kang, S., Scharmann, L. C., & Noh, T. (2005). Examining students’ views on the nature of science: Results from Korean 6th, 8th, and 10th graders. Science Education, 89 (2), 314-334.

Lederman, N. G. (1992). Students’ and teachers’ conceptions of the nature of science: A review of the research. Journal of Research in Science Teaching, 29, 331-359.

Lederman, N. G. & O’Malley, M. (1990). Students’ perceptions of tentativeness in science: Development, use, and sources of change. Science Education, 74 (2), 225-239.

Manassero, M. A. & Vázquez, A. (1998). Opinions sobre ciència, tecnologia i societat. Palma de Mallorca: Govern Balear, Conselleria d’Educació, Cultura i Esports.

Manassero, M. A. & Vázquez, A. (2001). Opiniones sobre las relaciones entre Ciencia, Tecnología y Sociedad. Tarbiya, 27, 27-56.

Manassero, M. A. & Vázquez, A. (2002). Las concepciones de estudiantes y profesores de ciencia, tecnología y su relación: Consecuencias para la educación. Revista de Ciencias de la Educación, 191, 315-343.

Manassero, M. A., Vázquez, A., & Acevedo, J. A. (2001). Avaluació dels temes de ciència, tecnologia i societat. Palma de Mallorca: Conselleria d’Educació i Cultura del Govern de les Illes Ballears.

Manassero, M. A., Vázquez, A., & Acevedo, J. A. (2003a). Cuestionario de opiniones sobre ciencia, tecnologia i societat (COCTS). Princeton, NJ: Educational Testing Service. Retrieved January 12, 2006, from: http://www.ets.org/testcoll/

Manassero, M. A., Vázquez, A., & Acevedo, J. A. (2003b, August, 19-23). Views on science-technology-society questionnaire: Categories and applications. Paper presented at 4th Conference of the European Science Education Research Association (ESERA), Research and the Quality of Science Education. Noordwijkerhout, Países Bajos. Retrieved January 4, 2006, from: http://www1.phys.uu.nl/esera2003/program.shtml

Manassero, M. A., Vázquez, A., & Acevedo, J. A. (2004). Evaluación de las actitudes del profesorado respecto a los temas STS: nuevos avances metodológicos. Enseñanza de las Ciencias, 22 (2), 299-312.

Martín Díaz, M. J., Gutiérrez Julián, M. S., & Gómez-Crespo, M. A. (2005, September, 7-10). Alfabetización científica. ¿Para qué y para quienes? ¿Cómo lograrla? Word presented at the VII Congreso Internacional sobre Investigación en la Didáctica de las Ciencias: Educación científica para la ciudadanía, Granada.

McComas, W. F. (1996). Ten myths of science: reexamining what we think we know about the nature of science. School Science and Mathematics, 96 (1), 10-16.

McComas, W. F. (1998). The principal elements of the nature of science: Dispelling the myths. In W. F. McComas (Ed.), The nature of science in science education (pp. 53-72). Dordrecht, Netherlands: Kluwer Academic.

Munby, H. (1997). Issues of validity in science attitude measurement. Journal ofResearch in Science Teaching, 34 (4), 337-341.

Ramsden, J. M. (1998). Mission impossible?: Can anything be done about attitudes to science. Internacional. Journal of Science Education, 20 (2), 125-137.

Rubba, P. A. & Harkness, W. L. (1993). Examination of preservice and in-service secondary science teachers’ beliefs about Science-Technology-Society interactions. Science Education, 77 (4), 407-431.

Rubba, P. A., Schoneweg, C. S., y Harkness, W. J. (1996). A new scoring procedure for the Views on Science-Technology-Society instrument. International Journal of Science Education, 18 (4), 387-400.

Ryan, A. G. (1987). High-school graduates’ beliefs about science-technology-society: IV. The characteristics of scientists. Science Education, 71 (4), 489-510.

Ryan, A. G. & Aikenhead, G. S. (1992). Students’ preconceptions about the epistemology of science. Science Education, 76 (6), 559-580.

Shrigley, R. L. & Koballa Jr., T. R. (1992). A decade of attitude research based on Hovland’s learning model. Science Education, 76 (1), 17-42.

Solbes, J. & Vilches, A. (1997). STS interactions and the teaching of Physics and Chemistry. Science Education, 81 (4), 377-386.

Solbes, J. & Vilches, A. (2002). Visiones de los estudiantes de secundaria acerca de las interacciones Ciencia, Tecnología y Sociedad. Revista Electrónica de Enseñanza de las Ciencias, 1 (2). Retrieved January 12, 2006, from: http://www.saum.uvigo.es/reec/.

Solbes, J. & Vilches, A. (2004). Papel de las relaciones entre ciencia, tecnología, sociedad y ambiente en la formación ciudadana. Enseñanza de las Ciencias, 22 (3), 337 348.

Solomon, J. (1994). Knowledge, values and the public choice of science knowledge. At J. Solomon & G. S. Aikenhead (Eds.), STS education: International perspectives on reform (pp. 99-110). New York: Teachers College Press.

Vázquez, A. (May, 1999). Innovando la enseñanza de las ciencias: El movimiento Ciencia-Tecnología-Sociedad. Boletín del Colegio de Doctores y Licenciados de Baleares, 25-35.

Vázquez, A., Acevedo, J. A., & Manassero, M. A. (2000). Progresos en la evaluación de actitudes relacionadas con la ciencia mediante el Cuestionario de Opiniones STS. In I. P. Martins (Coord.), O Movimento STS na Península Ibérica. Seminário Ibérico sobre Ciência-Tecnologia-Sociedade no ensino-aprendizagem das ciencias experimentais (pp. 219-230). Aveiro: Universidade de Aveiro. Retrieved January 12, 2006, from: http://www.campus-oei.org/salactsi/acevedo6.htm

Vázquez, A., Acevedo, J. A., & Manassero, M. A. (2005). Más allá de una enseñanza de las ciencias para científicos: hacia una educación científica humanística. Revista Electrónica de Enseñanza de las Ciencias, 4 (2). Retrieved January 22, 2006, from: http://www.saum.uvigo.es/reec/

Vázquez, A. & Manassero, M. A. (1995). Actitudes relacionadas con la ciencia: una revisión conceptual. Enseñanza de las Ciencias, 13 (3), 337-346.

Vázquez, A. & Manassero, M. A. (1996). La interacción entre ciencia, tecnología y sociedad: actitudes de los estudiantes. Taula, 25-26, 145-165.

Vázquez, A. & Manassero, M. A. (1999). Response and scoring models for the “Views on Science-Technology-Society” instrument. International Journal of Science Education, 21 (3), 231-247.

Vázquez, A., Manassero, M. A., & Acevedo, J. A. (2005). Análisis cuantitativo de Items complejos de option múltiple en ciencia, tecnología y sociedad: Escalamiento de Items. Revista Electrónica de Investigación Educativa, 7 (1). Retrieved January 14, 2006, from: http://redie.uabc.mx/vol7no1/contenido-vazquez.html

Vázquez, A., Manassero, M. A., & Acevedo, J. A. (2006). An analysis of complex multiple choice science-technology-society items: Methodological development and preliminary results. Science Education, 90 (4) 681-706.

Zoller, U., Donn, S., Wild, R., & Beckett, P. (1991a). Students’ versus their teachers’ beliefs and positions on science-technology-society oriented issues. International Journal of Science Education, 13 (1), 25-35.

Zoller, U., Donn, S., Wild, R., & Beckett, P. (1991b). Teachers’ beliefs and views on selected science-technology-society topics: A probe into STS literacy versus indoctrination. Science Education, 75 (5), 541-561.

Zoller, U., Ebenezer, J., Morely, K., Paras, S., Sandberg, V., West, C., Wolthers, T., & Tan, S. H. (1990). Goal attainment in science-technology-society (s/t/s) education and reality: The case of British Columbia. Science Education, 74 (1), 19-36.

Translator: Lessie Evona York-Weatherman

UABC Mexicali