Latane, B., Williams, K., & Harkins, S., Many Hands Make Light The Work: The Causes and Consequences of Social Loafing, JPSP, 37, 822-832.

"Two experiments found that when asked to perform a physically exerting tasks of clapping and shouting, people exhibit a sizable decrease in individual effort when performaing in groups as comapred to when they perform alone".

In a previous study over 50 years ago a German researcher had people alone and in groups pull on a rope attached to a strain guage to measure the pull force. Surprisingly, the sum of the individual pulls did not equal the total of the group pulls. Three people pulled at only 2.5X the average individual performance, and 8 pulled at less than 4x. The group result was much less then the sum of individual efforts.

This vilolates the notion that group effort and a sense of team participation leads to increased effort. Other social psychology studies would seem to suggest that individual performance does increase in a group for simple tasks with well-learned, dominant responses.

However, from a social pressure perspective, the social pressure to pull is further diluted for each person with increasing group size. From this perspective, if we assume people will work in proportion to the pressure they feel to do so, then bigger group sizes should result in less individual effort.

Experiment 1: Clap Your Hands and Shout Out Loud
The researchers chose clapping and shouting as a test exercise because people do these things in group setting and they are "maximizing, unitary, and additive".

They recruited subjects and put them in a semi-circle. Individually and in various group numbers, they asked people to either clap or shout as loud as they could. The measured noise level with a machine in sound pressure units.

Results
As with the rope pulling exercise, the level of noise increase with group size, but not in direction proportion. People averaged 3.7 dynes/sq cm alone, 2.6 in pairs, 1.8 in foursomes, and 1.5 in groups of six. There was no block effects (indicating tiredness or lack of practice).

The results don't fit with either Zajonc's social facilitation theory or evaluation apprehsion theory, They labeled this new drop in individual average performance in groups as "social loafing".

Other explanations
Another explanation is that individual effort does not decrease, but the group product decreases due to group inefficiency. However, another researcher duplicated the Ringelmann rope experiment and included a situation where people were blindfolded and led to believe others were pulling with them. They still observed a drop-off in performance with group size.

Another explanation may be acoustical -- the voices may cancel each other out or are not synchronized to be completely additive. The second experiment was designed to test this effect.

Experiment 2
Subjects were separated and put into rooms with headphones. Some were led to believe they were shouting alone, others believed they were shouting with a group. They repeated each trials with individual and group size options.

Results
The results were higher than the first experiment, but the trends were the same. Average individual performance decreased with group size. Groups ot two were 66% of capacity, groups of six at 36% of capacity. Even in situations where people thought they shouted together but in fact shouted alone, the "loafing effect" was present.

When performers belived one other person was yelling, they shouted 82% as intensly as alone. When they believe five others were yelling, they shouted at 74% as intensly as alone.

Discussion: The causes of social loafing
Three possible options are:
1. Attribution and Equity. Maybe subjects heard other's shouts as less than their own (because they were farther away than one's own voice) and felt the others were "slacking", leading them to back off on their shouting too. While the headphones should have made it very difficult to detect this kind of loafing, people could have come in with pre-conceived notions of how people slack off in groups.

2. Submaximal Goal Setting. Maybe subjects perceived there is some well-defined standard of loudness, and with more people they feel they can work less at it. The task would become an optimizing one than a maximizing one. (But the researchers felt that given the repeated exhortations to shout as loud as you can this couldn't be the case).

3. Lessened contingency between input and outcome. Maybe people felt they could "hide in the crowd" and avoid the negative consequences of slacking off, or maybe they felt "lost in the crowd" and unable to get their fair share of positive consequences for working hard. Only when performing alone can people be appropriately evaluated and rewarded.

Maybe by instiituting a payoff scheme that would avoid the benefits of a "free ride" in the group would this "social loafing" be reduced.

Social Loafing and Social Impact Theory
The three theories can be explained with Latanes social impact theory. "If a person is the target of social forces, increasing the number of other persons diminishes the relative social pressure on each person. If the individual inputs are not identifiable the person may work less hard. Thus if the person is dividing up the work to be performed or the amount of reward he expects to receive, he will work less hard in groups."

Social impact theory suggests that effort in group tasks should decrease as an inverse power function of the number of people in a group.

Other studies have also shown the "loafing effect" to hold in industrial production, bystander intervention, and participation in church activities.