Kallgren, Carl A., & Reno, Raymond R., & Cialdini, Robert B. (2000). A Focus Theory of Normative Conduct: When Norms Do and Do Not Affect Behavior. Personality and Social Psychology Bulletin, Vol. 26.
Submitted by Tracy Liu on Thu, 02/07/2008 - 12:38.
0
points
This paper illustrates how social norms direct individuals’ behaviors using three different experiments with diverse manipulation, which provides a fundamental solution about the controversial in the function of social norms.
The first experiment shows that the higher related-ness of passage to the antilittering norm with stronger arousal will decrease the percentage of littering. Furthermore, it demonstrates that only relatedness of the norm to the target norm was not enough to produce the effect if there is no focus. This result reveals that if we want to implement some regularity/norm on online communities, we need to make it obviously.
The second experiment finds that the “least littering would occur in the high norm-focus/two-handbill condition” (P1007), especially the effect of high norm-focus is important to induce regulated behaviors, which further strengths the effect of norm-focus to help social norm affect individuals’ behaviors.
The last experiment demonstrates that “less littering occurred both among participants who were internally focused and among participants who held strong antilittering personal norms” (P1009), further, if the internally focus is not induced; the strong antilittering personal norms could not direct individuals behavior to target norms. The implication of this result on online community design is that people might reduce their misbehaviors under special social norms if they are internally focused, but I haven’t figured out how to implement this purpose on online community design. How about publicize their opinion on special social norms?
One thing I am concerned with these studies is weather these findings on littering problem are also true on other problems. It seems that the authors didn’t give a comprehensive argument on this point.
Tracy: I wondered the same thing. How does one interpret the experimental results in a broader sense? Being more familiar with economic experimental techniques, I'm a bit wary of the author's generalization of their findings. I wonder how much of this is just a difference between experimental norms (!?!) in economics versus social psychology. On many levels, the philosophies are very different, e.g. economists NEVER use deception in experiments, while psychologists routinely do – this is not a judgement on the use of deception in experiments; these types of experiments would be very difficult/impossible to run without the element of deception.
On a broader scientific level, the internal/external validity seems to be problematic. I will say this, however: I was fascinated by the experimental design. What a fun experiment to run! The three treatments seem to have done a nice job of partitioning the different behavior spaces the experimenters were trying to test.
I think if we can just overcome this validity scope issue, the findings would be very useful when considering the design of an online community and how injunctive norms and personal norms affect behavior in participants.
The distinction between descriptive (is) and injunctive (ought) is interesting. I wonder to what extent the descriptive simply pre-injunctive and becomes transformed into a self-deployed injunctive.
I think the entire paper comes down to this quote:
"which... norms guides behavior is likely to be the one that is more focal at the time for action.they incorporate into those messages types of stimuli likely to be present in relevant behavioral settings."
And the key for design would be:
When a member of a community (online or otherwise) is about to make an action, make visible some element of the norm that is to influence the action.
-------------------------------------------------------- PHartzog@umich.edu
--------------------------------------------------------
The Universe is made up of stories, not atoms.
--Muriel Rukeyser
Submitted by Sean Munson on Sun, 02/10/2008 - 20:57.
0
points
I'm either not understanding the experimental design and operationalisms right, or I'm a bit concerrned about their first experimental design. At first I thought they considered arousal to be seeing the litter in the stairway (if they drew the exercise card), but then I see that it was physiological and measured as an increase in pulse from exercise (or are they referring to part of the deception here? I can't tell). They consider the physiological arousal to be a predictor, and they caused that physiological reaction while walking up and down the stairs. These are, if I'm not mistaken, the same stairs that were pre-littered? So was the physiological factor the predictor, or was it that they had 3 minutes of going up and down littered stairs to be exposed to an alternative norm (i.e., it's okay to litter)? My gut reaction would be the latter.
I'm also curious what the students thought the blue finger paint + petri dish part of the experiment accomplished (or did the experimenters just tell them they were measuring a generic physiological response?). Not that it necessarily important to the results, but I'm still curious, since it would make it easier for me to trust that the deception worked.
Submitted by Daniel Zhou on Mon, 02/11/2008 - 01:42.
0
points
The paper is based on previous research which concluded that "norm-focus procedures will be more useful in producing desirable behavior if the relevant norm is injunctive in character (specifying what most people approve/disapprove) rather than descriptive (specifying what most people do)." The main purpose of this work is to investigate the role of injunctive norm-focus procedures in the generation of socially desirable conduct. To do that, the authors conducted three experiments, which were summarized nicely by Tracy in the previous comment.
Actually, the main findings of the work can be summarized into one paragraph. If we only want to use the results in online community design, we don't have to go through the details of the experiments which are the main part of the paper (provided that we trust the authors and the journal). In fact, the previous readings are all like that. I was wondering how important it is to read the details of the experiments. Can we learn something from it? Or maybe we can just digest the results?
It seems to be a really typic psychology study in terms of the theory deriving, experiment design, and result analysis. We could learn the careful design for our e-community studies and others in HCI relevant things.
Also, I agree with above ideas on the external validaty of the studies in psychology, so much things are subjectve to this kind of doubt. For example, the the defined "norm" here, which is actually very easily adapting to the context, can be complication for the conclusion. That is this "norm" can be easier to be aroused than others, which may make the experimental predictors works better than on other "norms." For many behaviors in online communities, some can have very strong inclination, may be harder to manipulate.
Another thing I am wondering is how to measure "norm" in different contexts. What is the norms in a particular community? By intuition, we may anticipate that the strength of different "norms" on different contexts can be varing too.
Summary and one question about the universality of these results
This paper illustrates how social norms direct individuals’ behaviors using three different experiments with diverse manipulation, which provides a fundamental solution about the controversial in the function of social norms.
The first experiment shows that the higher related-ness of passage to the antilittering norm with stronger arousal will decrease the percentage of littering. Furthermore, it demonstrates that only relatedness of the norm to the target norm was not enough to produce the effect if there is no focus. This result reveals that if we want to implement some regularity/norm on online communities, we need to make it obviously.
The second experiment finds that the “least littering would occur in the high norm-focus/two-handbill condition” (P1007), especially the effect of high norm-focus is important to induce regulated behaviors, which further strengths the effect of norm-focus to help social norm affect individuals’ behaviors.
The last experiment demonstrates that “less littering occurred both among participants who were internally focused and among participants who held strong antilittering personal norms” (P1009), further, if the internally focus is not induced; the strong antilittering personal norms could not direct individuals behavior to target norms. The implication of this result on online community design is that people might reduce their misbehaviors under special social norms if they are internally focused, but I haven’t figured out how to implement this purpose on online community design. How about publicize their opinion on special social norms?
One thing I am concerned with these studies is weather these findings on littering problem are also true on other problems. It seems that the authors didn’t give a comprehensive argument on this point.
Internal validity versus external validity
Tracy: I wondered the same thing. How does one interpret the experimental results in a broader sense? Being more familiar with economic experimental techniques, I'm a bit wary of the author's generalization of their findings. I wonder how much of this is just a difference between experimental norms (!?!) in economics versus social psychology. On many levels, the philosophies are very different, e.g. economists NEVER use deception in experiments, while psychologists routinely do – this is not a judgement on the use of deception in experiments; these types of experiments would be very difficult/impossible to run without the element of deception.
On a broader scientific level, the internal/external validity seems to be problematic. I will say this, however: I was fascinated by the experimental design. What a fun experiment to run! The three treatments seem to have done a nice job of partitioning the different behavior spaces the experimenters were trying to test.
I think if we can just overcome this validity scope issue, the findings would be very useful when considering the design of an online community and how injunctive norms and personal norms affect behavior in participants.
Making Norms Visible
The distinction between descriptive (is) and injunctive (ought) is interesting. I wonder to what extent the descriptive simply pre-injunctive and becomes transformed into a self-deployed injunctive.
I think the entire paper comes down to this quote:
"which... norms guides behavior is likely to be the one that is more focal at the time for action.they incorporate into those messages types of stimuli likely to be present in relevant behavioral settings."
And the key for design would be:
When a member of a community (online or otherwise) is about to make an action, make visible some element of the norm that is to influence the action.
--------------------------------------------------------
PHartzog@umich.edu
--------------------------------------------------------
The Universe is made up of stories, not atoms.
--Muriel Rukeyser
confused by experimental design
I'm either not understanding the experimental design and operationalisms right, or I'm a bit concerrned about their first experimental design. At first I thought they considered arousal to be seeing the litter in the stairway (if they drew the exercise card), but then I see that it was physiological and measured as an increase in pulse from exercise (or are they referring to part of the deception here? I can't tell). They consider the physiological arousal to be a predictor, and they caused that physiological reaction while walking up and down the stairs. These are, if I'm not mistaken, the same stairs that were pre-littered? So was the physiological factor the predictor, or was it that they had 3 minutes of going up and down littered stairs to be exposed to an alternative norm (i.e., it's okay to litter)? My gut reaction would be the latter.
I'm also curious what the students thought the blue finger paint + petri dish part of the experiment accomplished (or did the experimenters just tell them they were measuring a generic physiological response?). Not that it necessarily important to the results, but I'm still curious, since it would make it easier for me to trust that the deception worked.
a brief summary
The paper is based on previous research which concluded that "norm-focus procedures will be more useful in producing desirable behavior if the relevant norm is injunctive in character (specifying what most people approve/disapprove) rather than descriptive (specifying what most people do)." The main purpose of this work is to investigate the role of injunctive norm-focus procedures in the generation of socially desirable conduct. To do that, the authors conducted three experiments, which were summarized nicely by Tracy in the previous comment.
Actually, the main findings of the work can be summarized into one paragraph. If we only want to use the results in online community design, we don't have to go through the details of the experiments which are the main part of the paper (provided that we trust the authors and the journal). In fact, the previous readings are all like that. I was wondering how important it is to read the details of the experiments. Can we learn something from it? Or maybe we can just digest the results?
Norm can be complicated
It seems to be a really typic psychology study in terms of the theory deriving, experiment design, and result analysis. We could learn the careful design for our e-community studies and others in HCI relevant things.
Also, I agree with above ideas on the external validaty of the studies in psychology, so much things are subjectve to this kind of doubt. For example, the the defined "norm" here, which is actually very easily adapting to the context, can be complication for the conclusion. That is this "norm" can be easier to be aroused than others, which may make the experimental predictors works better than on other "norms." For many behaviors in online communities, some can have very strong inclination, may be harder to manipulate.
Another thing I am wondering is how to measure "norm" in different contexts. What is the norms in a particular community? By intuition, we may anticipate that the strength of different "norms" on different contexts can be varing too.