Craving and Compulsive Involvement Scales

By Daniel Lende

In the previous post Wanting to Craving: Understanding Compulsive Involvement with Drugs, I wrote:

Different domains of subjective involvement can be linked to dopamine [function] – wanting more and more, the sense of an urge or push to use (often not a conscious desire), and the heightened focus on places and actions and times that lead to using. The scale I developed showed good internal consistency, adding support that these three senses of compulsive involvement are linked. If you want to know more about the scale, I have done a separate post detailing the compulsive involvement scale in both Spanish and English versions.

So this post gives you those scales! But first a little information. The scale in Spanish consists of eight items and shows good consistency. For the English version I have not fullly tested the scale, and have deliberately included more items there (a total of 13). One aim of future research (collaboration, anyone?) would be to test the items and hopefully winnow the size down.

So without further ado, here are the scales themselves:

Lende Craving-Salience General

Lende Deseo-Salience en Espanol

If you don’t want to click on any version, some typical statements in English include: (a) At times I have started to use and use without thinking about anything else; (b) At times I have felt a powerful impulse or urge to go use; and (c) At times using feels like you want more and more.

If you do want to use these scales, please contact me (Daniel Lende) at I’d like to hear more!

Results with the Spanish version originally appeared in my 2005 article Wanting and Drug Use: A Biocultural Approach to Addiction (Lende Wanting pdf). Here are excerpts from that article that relate directly to creating the original scale and its use in my study:

Construction of the Incentive Salience Scale
Ethnographic Results
The incentive salience scale was created by drawing on the results of the questionnaire and the first interview. The first step in this process was identifying the common dimensions related to wanting and seeking drugs in the adolescents’ descriptions. One important thing that emerged early in this review process was that the dimensions of wanting, shifts in attention, and behavioral engagement applied to more than just anticipating and seeking out drugs, the main focus of Robinson & Berridge’s theory. The dimensions of incentive salience applied to both seeking out and using drugs, leading to a wider focus on how drugs and drug use were salient to users.

Based on this wider view of salience, six common elements were then identified in the adolescents’ descriptions. First, one of the most typical ways of describing addictive experiences in Colombia was “querer más y más,” to want more and more drugs. During my ethnographic research, this emphasis on wanting—the Colombian’s summary description of what addiction was—took on more relevance as I realized the diversity in positive appraisals and “rewards” from substance use, ranging from “forgetting everything” to riding a skateboard better. Other ways used to describe this experience included “deseo” and “sentir ganas,” to feel a desire to consume drugs. Overall, the emphasis on wanting and desire provided a clear indication of the relevance of the incentive salience approach to understanding drug abuse in this population.

There were three commonalities across adolescent responses that related to shifts in attention while seeking out and using drugs. First, individuals emphasized “las sensaciones,” the sensations, they experienced while using drugs. An oft-repeated idea about these sensations was that they were very “present” in their minds. Second, substance abusers at times spoke of using drugs “sin darme cuenta”—“without realizing” or “without thinking about it,” a typical phrase to describe how individuals lost track of their amount of use. For example, one respondent narrated about having too many drinks at a party without realizing this and then suddenly seeing that he was stumbling around drunk with his head spinning. Third, when speaking about wanting to use drugs, users often said that this meant “de una vez,” “immediately” or “at once.” Abusers often told me that they could not wait, they had to go at once to use drugs, especially if “se presenta la oportunidad,” if the opportunity comes up.

Two final characteristics of the descriptions related to engagement in seeking out and using drugs. One phrase frequently used to describe addictive behavior was “dedicado a eso,” to be dedicated to that, in the sense of a person spending his or her time doing that and nothing else. Finally, the experience of use was often delineated in terms of “el momento,” the moment. Individuals spoke of being “metido en el momento,” of being into the moment, in a literal sense, being inserted into it.

Generation of Statements
At the same time that I was summarizing the main elements of drug experiences, I created a list of possible statements to include in the incentive scale by drawing on what the adolescents had written and said to me. Forty-eight statements were initially drawn out of the questionnaires and interviews based on the simple criterion of providing an idea for how drug use might be salient to adolescents there. Thirty two of these statements were then omitted based on two criteria: close similarity to another statement, and lack of a strong correspondence with the neurobiological framework.

The final scale contained eight items. These eight items were based on the ethnographic results covered above. Items that matched ideas of wanting, involvement, and attention but did not correspond to the common ethnographic themes were eliminated. Thus, the final scale had items that matched both the framework derived from the neurobiological review and the ethnographic research.

Using reliability analysis on all the items of the scale, the scale has an alpha of 0.949, well-above the normal cut-off of 0.8 (Nunnally and Bernstein 1994). Eliminating any item in the scale gave a lower overall alpha, indicating that all items were relevant to what the scale was designed to measure. A factor analysis extracted one significant factor (eigen value = 5.89, above the accepted cut-off of 1.0), accounting for 73.6% of the total variance. These results, that all items contribute to an internally consistent scale loading onto one factor, provide good evidence that the combination of theory and ethnographic results that guided the construction of this scale provides a useful way to develop a locally-appropriate measurement that addresses a biocultural variable.


Addicted Status: This dichotomous variable divided individuals into “addicted” (n=121) and “non addicted” (n=117) categories. In the “addicted” category, 89 individuals were from treatment and 32 from the school. In the non-addicted category, 10 individuals were from treatment and 107 from the school. Meeting any of the following three criteria indicated inclusion in addicted status: (1) reported using five times or more a day (the highest score), (2) reported trying to stop or cut back on substance use four times or more (the highest score), and (3) reported 17 or more on total problems related to substance use (averaging more than 2 reported problems over 8 different categories). All individuals not meeting at least one of these criteria were classified as non-addicted.

Incentive Salience: This eight-item scale provided a measure of how salient the experiences related to drug seeking and consumption were to the respondent (n=243, range 0 to 32, mean 15.67, SD 10.548). This variable was recoded so that 32 indicated the highest endorsement of incentive salience and 0 the lowest endorsement. Using a reliability analysis, this scale has an alpha = 0.949 in this data set.


Table 2: Logistic Regression Results For Addicted Status

                        Note: Results control for sex, age and research site.



Beta (b)



Sig. (p)


Lower 95% CI

Upper 95% CI

Incentive Salience








Drug Using Friends








Experiencing Armed Violence


















The promising thing about the incentive salience theory is that it offers a biologically-based theory that represents a significant alternative to the biomedical approach emphasizing the hard-wired reward of drugs and subsequent physiological withdrawal. The research presented here provides strong support for extending this animal-based model to the understanding of problems of substance abuse experienced by people. In particular, the epidemiological research demonstrated that the incentive salience score was a significant predictor of addicted status. Furthermore, the overall research confirmed the neurobiological framework of wanting, attention, and engagement developed to understand how incentive salience would be experienced by individuals.

In interpreting the results for incentive salience, the low odds ratio (exponent of beta) of 1.104 needs to be thought of in terms of the large range of this variable (0 to 32). Thus, for each one point increase in salience, the individual is 1.1 times more likely to be in the addicted category according to this study. However, the purpose of this study was not to produce a valid estimate of the risk that increasing incentive salience brings for substance abuse. These results provide strong support that incentive salience is a significant part of how we need to understand the problem of substance abuse, and further research is now needed to extend the assessment of incentive salience using a more formal epidemiological design.

3 thoughts on “Craving and Compulsive Involvement Scales

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s