pap 1
Ife PsychologIA; Volume 20 Number1, March 2012 Copyright ©2012 Ife Center for Psychological Studies/Services, Ile-Ife Nigeria.
229
EFFECTS OF GROUP SIZE AND EXPECTANCY OF REWARD ON
SOCIAL LOAFING
Philip C. Mefoh, PhD & Chinonso L. Nwanosike
Department of Psychology, University of Nigeria, Nsukka.
Abstract
There is a large body of evidence which have shown
that monitoring personal effort on group projects reduces social loafing effects, but as the world gets
more complex with several co-operative tasks there is
a need to explore other variables that would inspire
group, rather than individual performance. This
experiment re-examined the prediction that performing in large group would lead to social loafing behaviour,
the study also tested whether the promise of reward
would attenuate social loafing effect on a simple
experimental task. Forty (40) Igbo secondary school
students of south eastern Nigeria participated in the
experiment. They were aged 13 – 16 years (mean age = 14.53). Results indicated that performance was
significantly poorer in the group condition than in the
alone condition (p. <05); and reward significantly
attenuated social loafing effect in the group condition
(p. <05). These observations were discussed in relation to the prevailing challenge in team work.
Key words: Expectancy of reward, Group size, Max
Ringlemann, Individual performance, Personal effort.
Introduction
The Igbo speaking people of south eastern Nigeria knew that individuals work less hard in groups than when working alone. The
Igbo proverb, ―ewu oha nwe n‘anwu n‘aguu‖ (meaning that a goat
owned by many individuals usually dies of hunger), predict that
sharing work with other people reduce individual performance. This
phenomenon of individuals exerting less effort when they work in a group than when they work independently is labeled social loafing.
Social loafing is a pervasive characteristic of working in groups and
occurs in several different cultures (Gabrenya, Latane & Wang,
1983).
The first evidence of social loafing was demonstrated by a
French agricultural engineer, Max Ringlemann (Kravitz & Martin,
Ife PsychologIA; 20(1), March 2012
230
1986). Ringlemann asked participants to pull on a rope as hard as
they could. The participants pulled by themselves or with one, two,
or seven others. A sensitive gauge was used to measure how strongly the participants pulled the rope. Participants‘ efforts pulling on the
rope were less when they worked in a group than when they
performed the task alone. Groups of two pulled at only 95 percent of
their capacity and groups of three and eight sank to 85 percent and
49 percent respectively. That is, as more individuals pulled on the
rope, the individuals exerted themselves less. From these observations, Ringlemann observed that individuals perform below
their potential when working in a group (LaFasto & Larson, 2001).
Similarly, Latane, Williams and Harkins (1979) identified
social loafing among college students. Latane and colleagues had six
blindfolded college students sit in semi-circle. The students wore headphone that blasted sounds of people shouting into their ears.
The experimental task was to shout as loud as possible while
listening to the headphone noise. On some trials, the students
believed that the other five students were also shouting. At other
trials, they believe that they were either shouting alone or with one
other student. In reality, only a student was performing on all the trials. Consistent with the phenomenon of social loafing, when a
student thought one other person was shouting, the student shouted
82 percent as intensely as when alone, and when they believed all six
of them were shouting, they shouted 75 percent as intensely. As in
Ringlemann‘s study, output decreased with increased group membership, due to social loafing.
Since these early observations, social loafing has been
identified in numerous other studies (e.g., De Vita, 2001; Hardy &
Latane, 1988; Karau & Williams, 1993; Price & Harrison, 2006;
Weldone & Gargano, 1988). The phenomenon have also been
documented in a host of behaviours to illustrate a principle that is common in business, family, education, and in social gatherings that
harms the overall integrity and performance of a group by reducing
the level of output. The practical advice from most social loafing
research was that work groups should be designed so that each
individual‘s effort can be assessed independently of those of the group to prevent social loafing (Carron, Burke, & Prapavessis, 2004).
This advice was based on the observation that when individual
performance was monitored within the group situation the
individuals worked just as hard as they did when they worked alone
(Latane, 1981). That is, people working by themselves think they are
responsible for completing a task, when they work in groups;
Mefoh, P.; Effects of Group Size and Expetancy of Reward on Social Loafing
231
however, this feeling of responsibility diffuses to others and their
performance effort declines (Comer, 1995; Piezon & Donaldson,
2005). What happens when an individual‘s effort cannot be assessed
in a group activity? Many jobs today are very complex that only
group performance can be measured. Probably, one useful way to
attenuate social loafing effect on group projects (besides monitoring
personal effort) would be to develop a collectivistic culture. Not all
culture experience social loafing. Social loafing was more likely to occur in societies where the focus is on the individual rather than the
group (Gabrenya, et. al. 1985). In a study comparing individualistic
values to collectivistic values, Earley (1989) found that social loafing
occurred with managers with individualistic orientation, while there
was no such occurrence with managers whose orientation was focused on the group. In this study, the researchers predict that the
use of reward would enhance contributions in group product and
reduce social loafing.
The study has two simple objectives. The first was to
determine whether performance drop-off would occur in the group
condition and not in the alone condition; and the second was to examine whether reward will attenuate social loafing effect in group
condition. As many studies indicated that there was some degree of
social loafing within every group, it was predicted that performance
would be poorer in the group condition. Similarly, because it has
been demonstrated time after time in highly controlled learning experiments that reward increases the strength of a response and
increase its probability of being repeated in future (Eisenberga;
Armeli & Pretz, 1998), it was predicted that reward will reduce social
loafing effect in group conditions, and perhaps increase output.
Method
Participants
Participants for this experiment consist of forty (40) secondary school
students. All the participants were Igbo from the south eastern part
of Nigeria. The sex composition of the sample (17 males and 23 females) was representative of the student body from where the
sample was selected. The participants were selected by means of a
table of random numbers. Their ages ranges from 13 to 16 years (M =
14.53; SD = 2.93).
Ife PsychologIA; 20(1), March 2012
232
Figure 1
An example of the board matrix used in the experiment
Materials
The stimulus materials were 40 board matrices (5x5 ft.) similar to the
one shown in Figure 1. Each participant received 25 pieces of neatly
cut cardboard tiles and a board matrix. The blocks/cells spaces on
the matrices were in different shapes, some were square and others rectangle. The two (2) shapes were randomly mixed on the board
matrix, such that there were 15 square and 10 rectangular shaped
cells. Latency was measured using a stop-watch.
Procedure
This experiment adopted a pre-test post-test paradigm. Before the commencement of the experiment, participants were pre-
instructed that they would be tested twice. In the pre-test (alone)
condition, the entire participants received 25 pieces of carefully
marked cardboard tiles and a board matrix each. Thereafter, the
participants were randomly assigned into 2 groups and each group was tested in different class rooms. Participants were required to fix
the tiles to the spaces marked on the board matrix according to the
shape (either a square or a rectangle). They were shown examples of
how to fix the tiles, and were informed that only tiles that were
properly fixed unto the designated spaces on the board matrix would
be counted as correct. They were told to fix as many tiles as they can. Test time was 15s. After the trial, the tiles that were correctly fixed to
their specified positions were collected and counted. The participants
score was recorded and the tiles were returned back to the
participant for the next phase of the experiment.
The post-test (group) condition was the major interest of this experiment. In this phase, one of the 2 groups earlier created was
Mefoh, P.; Effects of Group Size and Expetancy of Reward on Social Loafing
233
referred to as the reward group and the other, the no-reward group.
The variable, reward, was manipulated by simple instruction.
Participants in the reward condition were given the following instruction:
―You will have to repeat the task again. This time, your individual scores would be added together with those of other members of your group to know which group
(referring to them and other participants in the next classroom), will do better than the other. If your group
does well in this task, each one of you would get an ink pen and an exercise book”.
Participants assigned to the no-reward condition were instructed
similarly, except that the statement promising reward was omitted.
The instruction given to the no-reward group stopped at: “… This time, your individual scores would be added together with those of other members of your group to know which group will do better that than the other”.
The instruction given to the two reward groups (reward vs.
no-reward) was designed to create an illusion of judgment. That is,
participants were led to believe that each individual‘s score on the task and those of other participants in the group would be added
together to obtain the group‘s total. During the post-test, the
researchers ensured that nothing untoward happened during the
experiment, and that all the participants performed the task strictly
to the rules. The groups‘ trials lasted for 15s. At the expiration of the test interval, the researchers gathered all the properly fixed tiles from
the board matrices and sorted them according to the participant who
fixed them. All the tiles used in this experiment were carefully
marked with identification numbers, making it easy for the
researchers to tell which participant fixed a particular tile. Thus, a
participant‘s score in the pre-test (alone) condition and his or her score on the post-test (group) condition were matched for
comparison.
After the pre- and post-test sessions, the researchers met
with all the participants in one of the classrooms used for the
experiment to explain the purpose of the test and why minimal deception was employed. All the participants in the study were given
an ink pen and an exercise book, regardless of whether the
participant was in the reward or no-reward condition. The gifts were
considered adequate compensation for the time the participants
spent in the research process.
Ife PsychologIA; 20(1), March 2012
234
Design/Statistics
This experiment adopted a composite design in which all 40
participants performed in both alone and group conditions. However, half that number was randomly assigned to the reward condition
while the other half were assigned to the no-reward condition.
Analysis of covariance (ANCOVA), because it provides a more
powerful test for pre-test post-test studies than difference score
(Dancey & Reidy, 2002), was adopted for data analysis.
Result FIGURE 2
Strict attention was paid to how participants placed the tiles
on the spaces demarcated on the board matrices. Only tiles that were
correctly placed were counted as correct. Data obtained in the study
were analyzed with analysis of covariance. The descriptive statistics showed that the number of tiles fixed by participants were more in
the pre-test (alone) condition (M=15.10; SD=2.27) than in the post-
test (group) condition (M=11.65; SD= 3.01). See Figure 2. An
evaluation of these scores showed that participants‘ scores on the
0
3
6
9
12
15
18
1A bar chart showing the mean number of tiles fixed in the alone and group
conditions
Alone Perf.
Group Perf.
Mefoh, P.; Effects of Group Size and Expetancy of Reward on Social Loafing
235
pre-test measure and their scores on the post-test measure differed
significantly, F (1, 37) = 4.82, P <.05). This result failed to reject the
first hypothesis that performance would be poorer in the group condition.
The result further indicated that more tiles were fixed by
participants assigned to the reward condition (M=13.05; SD= 2.72)
than those assigned to the no-reward condition (M=10.25; SD =
2.63). The two reward conditions (i.e., reward vs. no-reward) differed
significantly on the post-test measure after adjustment for the pre- test scores, F (1, 37) = 11.25, P <.05 (see Figure 3). This represented
an effect size of 0.23, showing that once pre-test scores were
partialled out, 23% of the variation in the number of tiles fixed on the
matrices can be accounted for by differing conditions of reward. This
result also failed to reject the second prediction that reward will reduce social loafing effect in the group condition.
FIGURE 3
A related t-test was used to compare the scores obtained by
participants assigned to the post-test/reward condition with the
scores they obtained in the pre-test measure. If the mean score obtained by participants in the post-test/reward condition was
meaningfully greater than the mean score obtained by the same
participants on the pre-test measure, it would mean that reward
increased productivity. This was not the case, instead the number of
0
3
6
9
12
15
18
1A bar chart showing the mean number of tiles fixed in the reward and no reward
conditions.
Reward Condition
No Reward condition
Ife PsychologIA; 20(1), March 2012
236
tiles fixed by participants in the post-test/reward condition were
fewer (261) compared to their performance in the pre-test measure
(307). Analysis of these data showed that the difference between the group‘s post-test and pre-test scores was 2.15; and the t-value (3.25)
has an associated p-value of p<.004. This finding indicated that
although reward attenuated social loafing behavior, it not increase or
improve productivity.
Discussion This study sought to determine whether performance drop-off
would occur in the group condition and not in the alone condition;
and to examine whether reward will attenuate social loafing effect in
the group condition and probably increase productivity. Result
obtained in relation to the first objective showed that participants worked less hard in the group condition than in the alone condition.
This observation was consistent with previous studies (Karau &
Williams, 1995; Kravitz & Martin, 1986; Price & Harrison, 2006;
Weldon & Gargano, 1988), that social loafing engenders negative
consequences that affect groups. Like in most of those studies the
observation made in this experiment can be related to the theory of diffusion of responsibility. People working alone think they are
personally responsible for task outcome; but when they work in
groups, this feeling of responsibility diffuses to others. Rothwell
(2004) argued that large group sizes can cause individuals to feel lost
in the crowed. In this experiment, participants were led to believe that their scores on the post-test (group) condition would be added
together to those of nineteen other participants to arrive at the
group‘s total. Because the group size was too large, it was possible
that participants, especially those driven by their individuality felt
that their contribution will not be recognized, and this lowered their
motivation. When a group becomes very large, some people often feel that their efforts are not needed or will not be recognized (Kerr,
1989).
The second objective that reward will attenuate social loafing
effect in participants who were expecting reward was confirmed. The
performance mean score of participants assigned to the reward and no-reward conditions differed significantly in favour of participants
who were promised reward. The observation was consistent with
literature (e.g., Eisenbarger, Armeli & Pretz, 1998), that reward
control behaviour and facilitate performance. However, when an
individual believes that compensation was not allotted equally among
group members, the individual withdraws his/her individual efforts.
Mefoh, P.; Effects of Group Size and Expetancy of Reward on Social Loafing
237
In this experiment, because participants believed that distribution of
compensation would be equitable, they anticipated reward contingent
on group goal. This disposition probably reinforced the participants‘ desire to pursue group goals in order to benefit the group. That is,
participants in the reward condition were motivated to pursue group
goals and they had hoped that other group members will also
contribute to the group‘s performance.
The post hoc result showed that reward did not increase
productivity. Although reward significantly reduced social loafing effect, it failed to increase performance/productivity on the
experimental task. In the experiment, the performance of the
participants assigned to the post-test/reward condition was poorer
compare to their pre-test scores. It can be argued that when reward
can be obtained just by belonging to a group, group members sometimes become less committed to group goals, have lower levels
of performance, set lower goals for their own achievement and
constitute group whose performance suffers. Therefore, to use reward
to improve productivity, it should be merged with guiding individuals
to exhibit loyalty to group objectives. When people know the goal,
know how far they need to go and where the competition is, they are more inclined to work towards the goals than if they did not have
that knowledge.
In conclusion, this experiment collaborated with many
previous studies that social loafing was a pervasive characteristic of
working in groups. However, social loafing is not inevitable. Latane (1981) demonstrated that when individual performance was
monitored within the group situation, social striving rather than
social loafing occur. Creating structures where individual
contributions are observable, measurable and known to all team
members attenuate social loafing effect. This technique had remained
the best known solution to curbing social loafing effect in group settings. But because the contemporary World is becoming more
complex by each passing day, and because several tasks are co-
operative in nature, personal effort may be difficult, if not impossible
to monitor in such settings. It was this realization that caused the
researchers to examine the effect of reward in attenuating social loafing effect. The experiment demonstrated that reward attenuated
social loafing behavior on the experimental task, but did not quite
increase productivity. The finding tends to suggest that reward is
only but a single variable that may influence social loafing
behaviour/effect. This calls for more research, it may require an
integrative framework of analysis of the reward – other variable(s)
Ife PsychologIA; 20(1), March 2012
238
interactions to understand the relationship between reward and
productivity in social loafing research.
Mefoh, P.; Effects of Group Size and Expetancy of Reward on Social Loafing
239
References
Brooks, C. & Ammons, J. (2003): Free riding in group projects and
the effect of timing, frequency and specificity of criteria in peer assessment. Journal of Education for Business, May- June.
Carron, A. Burke, S. & Prapavessis, H. (2004): managing groups and teams social loafing. Journal of Applied Sport Psychology, vol. 16, 41-58.
Comer, D.R. (1995): A model of social loafing in real work group. Human Relations, vol. 48, 647-667.
Dancey, C.P. & Reidy, J. (2002): Statistics without Maths for Psychology. Essex: Pearson Education.
De Vita, G. (2001): The use of group work in large and diverse business management classes: Some critical issues. The International Journal of Management Education, vol. 1 (3), 26-
34.
Earley, P.C. (1989): Social loafing and collectivism: A comparison of
the United States and the people‘s Republic of China. Administrative Science Quarterly in Business, vol. 34, 565-
581.
Eisenberger, R. Armeli, S, & Pretz, J. (1998): Can the promise of reward increase creativity? Journal of Personality and Social Psychology, vol. 66 (6), 1116-1127.
Gabrenya, W.K., Latane, B & Wang, Y. (1983): Social loafing in cross- cultural perspective: Chinese on Taiwan. Journal of Cross- Cultural Psychology, vol. 14, 368-384.
Gabrenya, W.K. Jr, Wang, Y.E, & Latane, B. (1985): Social loafing on
an optimizing task: Cross-cultural differences among Chinese and Americans. Journal of Cross –Cultural Psychology, vol. 16,
223-242.
Hardy, C.J. & Latane, B. (1988): Social loafing in Cheerleaders: Effects of team membership and competition. Journal of Sport and Exercise Psychology, vol. 10, 109-114.
Karau, S.J. & Williams, K.D. (1993): Social loafing: A meta-analytic review and theoretical integration. Journal of Personality and Social Psychology, vol. 65, 681-706.
Kerr, N.L. (1989): Illusions of efficacy: The effects of group size on
perceived efficacy in social dilemmas. Journal of Experimental Social Psychology, vol. 25, 287-313.
Kravitz, D. A. & Martin, B. (1986): Ringlemann rediscovered: The original article. Journal of Personality and Social Psychology, vol. 50 (5), 936-941.
Ife PsychologIA; 20(1), March 2012
240
LaFasto, F. & Larson, C. (2001): When teams work best. Thousand
oaks, CA: sage Publications. Latane, B. (1981): The Psychology of social impact. American
Psychologist, vol. 36, 343-356.
Latane, B., Williams, K. & Harkins, S. (1979): Many hands make light the work: Causes and consequences of social loafing. Journal of Personality and Social Psychology, vol. 27, 822-832.
Piezon, S.L., & Donaldson, R.L. (2005): Online groups and social loafing: understanding student- group interaction. Online
Journal of Distance Learning Administration, vol. 8 (4).
Retrieved online July 7, 2007 http://www.westga.edu /adistance/ojdla/winter 84/ piezon 84 htm.
Price, K.H & Harrison, D.A. (2006): Withholding inputs in team
context: Member composition, interaction process, evaluation structure, and social loafing. Journal of Applied Psychology, vol. 91 (6), 84 – 94.
Rothwell, D. (2004): In mixed company. California: Thomson
Wadsworth. Weldone, E. & Gargano, G.M. (1988): Cognitive loading: The effects of
accountability and shared responsibility on cognitive effort. Personality and Social Psychology Bulletin, vol. 14, 159 - 171.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.