Teaching the Analysis of Titles:
Dependent and Independent Variables in Research Articles

Dr. Diane Nahl
and
Dr. Leon James

Contents

Titles of Experimental Reports in the Social Sciences
Titles from the Physical and Biochemical Sciences
Teaching the Analysis of Titles
Conclusions
References

The titles of scientific articles that are primarily reports of experiments usually contain information about the cause-effect relation investigated. "The Effect of Alcohol on Driving Impairment" follows the common pattern of "The Effect of IV on DV" where IV refers to the independent variable or treatment condition, and DV refers to the dependent variable or treatment effect. Understanding the structure of titles is an important bibliographic skill. Teaching the analysis of titles allows librarians to contribute to science education and advanced reading skills in a unique and interesting way.

To librarians and scientists titles in publications are important elements in the organization and retrieval of scholarly information. Title-scanning is a regular professional activity. We search for titles in places such as catalogues, periodical indexes, bibliographies, references, and tables of contents of edited books, reports, and proceedings. The titles of scientific articles typically give details about the research design of the experiment. This is understandable since other researchers are keenly interested in knowing about experiments that relate directly to their research problem. The most basic information about an experiment is the identification of the variables or measurements that were made and the conditions under which they were obtained. To researchers this is known as identifying the dependent and independent variables. Authors of research articles try to convey this information in the title.

To the untrained reader this information is not immediately apparent from the title, but the skill can be learned with a little practice since it involves natural language abilities. We will explain several examples in what follows.

Titles of Experimental Reports in the Social Sciences

All scientific experiments share one central design characteristic, namely, the attempt to identify a cause-effect relation. For example, an article bearing the title Drinking Alcohol and Driving Impairment clearly advertises itself as a discussion on how drinking alcohol affects driving. Drinking alcohol is the cause and impairment in driving is the effect. The phrase "Drinking Alcohol" is the name of the Independent Variable (IV) while "Driving Impairment" is the name of the Dependent Variable (DV). A "variable" is anything that has been measured or quantified. In this case, if we assume that the article is reporting an experiment, we can infer that some treatment was administered (IV), such as the measured intake of alcohol, and some driving test score was obtained for every subject (DV).

It helps to remember that experimenters think in terms of DVs and IVs just as librarians think in terms of subject headings. It is a natural and automatic focus, though it takes some practice to attain i~. Inspecting the titles of scientific articles reveals that the majority of those that report experiments are based on the IV-DV format. Decoding this structure in titles is sometimes a problem because there are a variety of stylistic ways of expressing this simple underlying cause-effect relationship. For instance, it takes a little more processing to decode the title Alcohol Consumption Rates and Highway Fatalities since the phrase "Alcohol Consumption Rates" must be translated into "Drinking Alcohol" and "Highway Fatalities" must be paraphrased as "Driving Impairment." Note that variables in titles are also keywords in a field.

Some frequently used formats for titles of experimental reports are summarized by the following sentence frames:

The Effect of IV on DV

The Role of IV on DV (or in DY)

DV as a Result of IV

IV and DV (or DV and IV)

IVl and IV2 as determinants of DVl, DV2, and DV3

DV Characteristics of IV Systems

DV1 and DV2 in IVl

The IV always refers to an experimental treatment or intervention. The DY always refers to the measured consequences of the intervention or manipulation. Take as an example the title of the following reference: Amir, Y. (1976). The Role of Intergroup Contact in Change of Prejudice and Ethnic Relations. In P.A. Katz (ed.), Towards the Elimination of Racism. New York: Pergamon. This is obviously a chapter in an edited book. The title contains the keywords "Intergroup Contact," "Prejudice," and "Ethnic Relations." We would expect these to be listed in a book index. The title implies that the chapter will discuss the cause-effect relation between greater interracial contact and a consequent lessening of racial prejudice. The phrase "Intergroup Contact" is to be seen in relation to racial prejudice so that it is recognized as a frequently advocated method for the reduction or elimination of racism. It is thus the IV manipulated by the experimenter to produce "Change of Prejudice" which is thus the DV. The element "Ethnic Relations" can be interpreted as either a second DV or an extended reference to prejudice. This title thus fits the format, The Role of IV in DVl and DV2.

Note that the expression "The Role of" is optional. If you drop it from the title, it is still there implicitly as in the following:

Anderson, L.R. & Blanchard, P.N. (1982). Sex Differences in Task and Social-Emotional Behavior. Basic and Applied Social Psychology, 3. 109-139.

Here the DV is "Task and Social-Emotional Behavior" which actually implies two DVs, namely "Task Behavior" and "Social-Emotional Behavior." These appear to refer to performance scores ("Behavior") on executing a "Task" and on a rating obtained on the subjects' reactions ("Social-Emotional',). The IV is "Sex Differences" or "Sex" (i.e., gender). In other words the study investigates performance differences (DY) in men vs. women (IV). According to the design of the study, the IV of Sex (or gender, i.e., being a man or woman) causes the difference in performance (DV). The experimenter used the variable of Sex (or gender) to select subjects, choosing men and women by some sampling method (IV) and administering a common treatment to both groups (DV). Here we know that the treatment was to have each subject perform a task (not specified in the title) while their social or emotional reactions were noted and given a score. The scores (DV) of the men and women (IV) were then compared. These details can be specified from the title alone if the reader has some elementary knowledge of scientific research design and statistical analysis.

The following example shows that titles of scientific research reports presuppose knowledge of thesaurus entries in a specialized field or active research area:

Arkin, R.M., Appelman, A.J., and Burger, J.M. (1980).

Social Anxiety, Self-Presentation, and the Self-Serving Bias in Causal Attribution. Journal of Personality and Social Psychology, 38, 23-35.

"Causal Attribution" is a research topic dealing with the study of how people identify the causes of the events they perceive. "Self-Serving Bias" refers to people's tendency to be partial to themselves when assigning responsibility for blame. The phrase "the Self-Serving Bias in Causal Attribution" thus fits the pattern, The DV in IV. The IV is the experimental treatment of exposing subjects to a social situation in which they are called upon to assign blame for an event ("Causal Attribution"). The DVs are "Social Anxiety," "Self-Presentation," and "Self-Serving Bias;" the first two are personality test scores obtained by every subject, and the third is the score obtained in the causal attribution task. The study is thus about how the personality scores (DV1 and DV2) of subjects are related to their self-bias scores (DVj) in a situation where they are asked to judge who should be blamed for an event (IV).

Titles from the Physical and Biochemical Sciences

Though all experimental sciences share a common origin and central focus, one can appreciate that treatment conditions and measurements would differ in kind between human subjects and physical objects. This difference needs to be taken into account in the reasoning process whereby we infer which element in the title is the DV and which the IV. Consider this entry in The Engineering Index Annual - 1983: 043478 Heat Transfer by Natural Convection in a Vertical Porous Layer. "Heat Transfer" is a DV referring to the measurement of the amount of heat displaced from one place to another. Since "by Natural Convection" specifies the method by which heat is displaced, it is the IV, i.e., the treatment manipulation chosen by the experimenter in setting up the design of the experiment. The title further specifies a feature of the IV, namely that convection was generated "in a Vertical Porous Layer."

An article appearing in the Journal of Microscopy, Vo1.134(2), May 1984, pp.203-211, has the following title:

Liquid Propane Jet-Freezing, Freeze-Drying and Rotary Replication of the Cytoskeleton and Plasma Membrane Associated Structures in Human Monocytes

Here we are told that certain cells ("Human Monocytes") were frozen by various methods, and that the resulting effects on membrane structures were studied. The IV is the method of freezing ("Liquid Propane" vs. "Freeze-Drying" vs. "Rotary Replication"). The DVs are the measures made on the cell membranes. These are not specified in the title and we would expect that researchers in the field would know what they are. The frame of this title can be rendered, IVl, IV2, and IV3 of DV (or in DV). The December 20, 1982 issue of Physical Review Letters, Voi.49(25), p.1861 contains a one-page article with the following title: Mechanical Measurements of the Melting Transition in Thin Liquid-Crystal Films The DVs are "Mechanical Measurements of the Melting Transition" while the IV specifies the method or medium used to generate the melting scores ("in Liquid-Crystal Films.") The Abstract of the article gives details about the type of mechanical properties that were measured (DVs) during melting of the substance (IV). The title format is thus, DVs in IV.

Teaching the Analysis of Titles

Teaching about titles offers librarians the opportunity to contribute to science education in a unique and interesting way. A lecture or course might proceed in three phases. First, introducing the topic of "Titles and Its Properties" This may be restricted to certain materials like experiments, songs of a particular period, or it may be a discussion on titles in general. Second, demonstrating these properties by analyzing the elements of various titles and making explicit the IV-DV format. Third, giving practice exercises to the students and discussing the correct solutions. These steps are illustrated in what follows.

Step 1: Introduction.

Surface and Underlying Structure of Titles. Everybody who talks uses titles. For example, when asked about a trip, a person replies, "It was a fun trip. But I'm glad to be back. There is nothing like home." In this reply there are several titles which one might use for a story about the trip: Fun Trip or Glad to Be Back or There is Nothing Like Home. Titling a book, painting, song, or experience forms part of one's natural language abilities.

Titles, like phrases and sentences, are spontaneously built up by means of syntactic and semantic relations. Linguists like Noam Chomsky at MIT have shown that titles or sentences have a surface literal meaning and an underlying implicit meaning.1 He illustrates this property with the sentence, They are cooking apples, pointing out that the sentence is two-way ambiguous. Thus, either some people are in the kitchen preparing an apple pie, or someone is at the supermarket deciding that the apples are the cooking kind. Chomsky shows that the ambiguity of the "surface structure" is due to the presence of two different "underlying structures" -- one dealing with kitchen activities, the other with supermarket activities. According to linguists, a particular sentence or phrase is "generated" by a deep structure meaning or intention that is mapped onto a variety of stylistically different surface structures. Note that natural language allows us to choose a variety of ways to express an underlying meaning in a surface string of words. This is called paraphrasing. Try paraphrasing the title Last Summer at Disneyland and you'll see there are several options for communicating the same idea. The listener will still get the same idea even though you can express it in different ways.

Paraphrasing Titles. The ability to paraphrase is an essential component of library research. When searching, we constantly paraphrase or translate titles, subject headings, keywords, and search terms into our own thinking to determine whether they are relevant to our information need. For example, The Bagdad Thief is equivalent to The Thief of Bagdad. Similarly, Mice Resistance can be paraphrased as Resistance in Mice. This aspect of titles and subject headings is very important to retrieving information from alphabetically arranged databases such as the library catalog, or indexes such as the Library of Congress Subject Headings (LCSH). When you're looking for "Resistance in Mice" in an index you'll discover that it is not under "R" but under "M" for "Mice, Resistance." Flexibility in paraphrasing keywords, search terms, and titles is a key element that determines the success of searching.

Decoding and Encoding Titles. We "generate" phrases and titles in response to an inner need or intention. This is called encoding information, that is, putting into a standard code one's inner meaning or personal intention in a situation. Once coded the information is in its external or surface structure. Generating a title is thus a process of communicating to others some underlying inner intention by encoding it into a surface string of words. When the title is read or heard the surface structure needs to be analyzed or decoded back into its underlying structure that corresponds to the intention that generated it. Encoding and decoding titles is the main activity in dialog and communication.

The Purpose of Titles. Psychologist B.F. Skinner of Harvard views sentences and titles as verbal responses or "operants" motivated by a person's desire to influence the behavior of the listener.2 For instance, when you want someone to move out of the way you can say "Excuse me" and more often than not the person will move out of your way. "Audience control" refers to the reverse influence of the audience upon us whereby we adjust the style of our response to the characteristics of the audience or situation. For example, titling the experiences we've had at a reception is done differently when reporting it to a casual acquaintance versus a spouse. Similarly, when authors make up a title, say for a research proposal to a funding institution, they are motivated to attract the attention and interest of the reviewers.

Scientists engaged in an active research area are very eager to know about the experiments that other scientists carry out on the same problem. To facilitate this communication need groups of specialists in an area get together at meetings and conventions and read papers to each other. They also read each other's articles written in journals, magazines, newsletters, and reports of proceedings, as well as relying on reference sources such as periodical indexes, abstracts, and annotated bibliographies. These databases arrange the abstracts or titles of the reports by subject so that the user can retrieve articles by looking up their favorite subject headings or keywords. All of this information exchange activity is made possible because of the nature of titles. Linking titles into a network or sub-group is the essential purpose of library research. Titles form the connecting point between writer and reader, or between speaker and listener, or between one scholar and another, or between the artist and the patron. The march of ideas across time, place, and generation progresses upon the back of titles.

Step 2: Demonstrating The Analysis of Titles.

The IV-DV Format. You can view titles as abbreviated sentences. The purpose of analysis is to reconstruct the whole by looking at its parts separately. Let us take actual titles of published research articles in different fields.

Post, G., D.V. Power and T.M. Kloppel. 1974. Survival of Rainbow Trout Eggs After

Receiving Physical Shocks of Known Magnitude, Trans. Am. Fish Soc., 103:711-716.

This title almost makes a full sentence and is easy to decode. The DV is "Survival of Rainbow Trout Eggs" and the IV is "After Receiving Physical Shocks." In other words, an experiment is reported in which eggs are shocked (IV or the treatment condition) and their survival rate measured (DV or the treatment's effects). The sentence frame is DV After IV.

A common format to express a cause-effect relation is The DV in the IV (or on the IV) such as The Thief of Bagdad or Cat on a Hot Tin Roof. The specification "of Bagdad" or "on a Hot Tin Roof" is similar in function to the IV in an experiment because the specification selects from other choices such as "of Paris" or "on a Hot Asphalt Road." The elements "The Thief" and "Cat" function as DVs in experiments because they allow us to assess something about the IV (Bagdad and Hot Tin Roof). Notice the same pattern in the following scientific citation:

Evans, D.H. 1975. Ionic Exchange Mechanism in Fish Gills,Comp. Biochem.

Physiol., A: Comp. Physiol., 51:491-495.

"Ionic Exchange Mechanism" is the DV and "in Fish Gills" is the IV.

Journals, magazines, and newsletters also contain articles that are not reports of experiments but discuss information that is relevant to the professional and theoretical concerns of scientists. Here are two examples from the medical and nursing literature:

Turnbull, C. Adult Learning Theory: Its Applications to the Education of Midwives, Aust. J. Adv. Nurs. 1986 Jun.-Aug.; 3(4): 51-9.

Twomey, L. Physiotherapy and Health Promotion (editorial) Physiother. Pract. 1986 Dec.; 2(4): 153-4.

The discovery of cause-effect relations is at the heart of science. Therefore this concern will permeate nearly all professional concerns of scientists and practitioners. These two references clearly do not report an experiment yet the interest in cause-effect relations is apparent. In the first example, a nursing journal article discusses how learning theory can help in the training of midwives. Since many things can conceivably help train midwives, the choice of "Adult Learning Theory" functions as an IV or treatment condition, and "the Education of Midwives" then functions as the DV or the treatment effect. In the second example, an editorial discusses the relation between physiotherapy and health. On the surface, these two subjects appear as parallel topics that are interlinked. However, examining the underlying structure, or inner intention that generated it, we notice that physiotherapy is thought of as promoting health in a cause-effect relation. This means that "Physiotherapy" is the IV while "Health" or "Health Promotion" is the DV.

Step 3: Exercises.

We have experimented with two types -- decoding and encoding. The three types of exercises we tried for decoding titles may be described as follows:

(a) Give titles and have students identify the keywords or subject headings. They may

use LCSH or a thesaurus to verify their answers;

(b) Give titles and have students identify the DV and IV. Referring to the article Abstract helps verify the solutions;

(c) Give titles or full citations and have students group them by field or research area.

Looking at the source of the references and their place of publication helps verify the

solutions.

We tried two types of exercises for encoding titles:

(a) Give subject headings, keywords, or "variables" and have students make up titles

that fit the DV-iV pattern. Read the various solutions out loud to the class;

(b) Give an Abstract and have students make up a title for the article. Verify solutions

by comparing to the actual article title.

Conclusions

Teaching the analysis of titles may become an important aspect of bibliographic instruction. It is an excellent opportunity for librarians to contribute to the science education of their patrons. Learning to identify the DV-[V relation in scientific and professional articles helps one gain a new perspective on the scientific method. It also teaches a higher, more advanced form of reading skills. The analysis of titles renders explicit what otherwise might remain implicit and unnoticed. Understanding titles increases one's free access to information. Through teaching title analysis library and information science has a proper role to play in both science education and teaching reading skills.

References

1 Noam Chomsky, Aspects of the Theory of Syntax 1965 (Cambridge: MIT Press).

2 B.F. Skinner, Verbal Behavior 1957 (New York: Appleton-Century-Crofts).