Gender Differences in Christmas Gift-Giving

Abstract

In Christmas gift giving, the value of a gift is broadly representative of the strength of the relationship between gift-giver and recipient. In this study we examine the effects of relatedness, social proximity, and donor gender on self-reported patterns of Christmas gift-giving. As is consistent with kin selection and social network theory, respondents bought gifts of greater value for more closely related kin, and individuals in more proximate social network layers. Although men and women spent similar amounts on gifts for kin and close network layers, there was a striking gender difference in spending on friends and the most distant network layer, with women spending significantly more.

Full text

Gender Differences in Christmas Gift-Giving
Mark Dyble and Abram J. van Leeuwen
University College London R. I. M. Dunbar
University of Oxford
In Christmas gift giving, the value of a gift is broadly representative of the strength of
the relationship between gift-giver and recipient. In this study we examine the effects
of relatedness, social proximity, and donor gender on self-reported patterns of Christ-
mas gift-giving. As is consistent with kin selection and social network theory, respon-
dents bought gifts of greater value for more closely related kin, and individuals in more
proximate social network layers. Although men and women spent similar amounts on
gifts for kin and close network layers, there was a striking gender difference in
spending on friends and the most distant network layer, with women spending signif-
icantly more.
Keywords: altruism, friendship, gender differences, gift-giving, kinship
As well as an important form of economic
exchange, gift-giving is a mechanism by which
individuals initiate and maintain social relation-
ships in many human societies. The importance
of gift giving is reflected in the large amount of
work on the subject within the fields of anthro-
pology (Mauss, 1954;Sherry, 1983), sociology
(Caplow, 1982;Cheal, 1988), and marketing
(Laroche, Saad, Kim, & Browne, 2000). More
recently, gift giving has also been examined
within the framework of evolutionary psychol-
ogy (Iredale, Van Vugt, & Dunbar, 2008;Saad
& Gill, 2003), building on a body of theoretical
and empirical studies of investment strategies in
nonhuman animals (Barrett & Hanzi, 2001;
Colmenares, Zaragoza, & Hernandez-Lloreda,
2002;Vahed, 1998), and representing a part of
the growing interest in the evolutionary basis of
human consumption (Saad, 2007,2011).
A number of studies have tested evolutionary
predictions in the context of Christmas gift-
giving, finding significant effects of sex and
donor-recipient relatedness on gift-giving pat-
terns. In an analysis of self-reported gift values,
Mysterud et al. (2006) found that gifts of greater
value were given to more closely related indi-
viduals, that women exchanged gifts with a
larger number of friends than men, and that
women spent more on gifts than men did. In
contrast, in an analysis of intended spending
patterns, Jonason et al. (2009) found that men
planned to spend more on Christmas gift giving
in general than women did, although women
planned to spend more on friends and family
than men. In this study we aim to further ex-
plore the effects of donor-recipient relatedness
and sex on patterns of Christmas gift-giving.
We also aim to extend existing work by exam-
ining the effect of the strength of the social
relationship between donor and recipient on gift
value. In line with previous work on altruism in
social networks (Curry, Roberts, & Dunbar,
2013), we predict that individuals will spend
significantly more on gifts for recipients in more
proximate social network layers.
Method
Participants and Procedure
Data were gathered via an online question-
naire in which respondents were asked to pro-
vide information about all Christmas gifts they
had bought, including (a) the cash value of the
gift in British pounds (GBP), (b) their relation-
This article was published Online First September 22,
2014.
Mark Dyble and Abram J. van Leeuwen, Department of
Anthropology, University College London; R. I. M. Dun-
bar, Department of Experimental Psychology, University of
Oxford.
Correspondence concerning this article should be addressed
to Mark Dyble, Department of Anthropology, University Col-
lege London, 14 Taviton Street, London WC1H 0BW, United
Kingdom. E-mail: mark.dyble@cantab.net
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Evolutionary Behavioral Sciences © 2014 American Psychological Association
2015, Vol. 9, No. 2, 140–144 2330-2925/15/$12.00 http://dx.doi.org/10.1037/ebs0000022
140

ship with the gift recipient, and (c) the fre-
quency of contact with the recipient. Addition-
ally, respondents provided their own age and
sex. In total, 99 participants fully completed the
survey, of which 38 were men and 61 were
women. The mean age was 23.6 years (SD ⫽
2.6), range 19–31. Across the 99 participants,
562 gifts were given.
Frequency of contact was recoded into a net-
work layer variable according to a scheme used
in previous studies (Curry et al., 2013;Roberts,
Wilson, Fedurek, & Dunbar, 2008) where
weekly contact represents layer 1 (the ‘support
network’), monthly contact represents layer 2
(‘sympathy network’), and yearly contact layer
3 (the ‘active network’). The remainder form
part of the inactive or ‘acquaintances’ network.
Relatives were categorized according to biolog-
ical relatedness (r), such that, for example, par-
ents and siblings are r⫽.5, aunts, uncles, and
grandparents are r⫽.25, and first cousins are
r⫽.125.
Analysis
Because the data had a hierarchical structure,
with gift recipients being nested by donor, a
multilevel model was used to analyze the data
(also known as a hierarchical linear model).
Although gift value between 1 and 100 GBP
was provided to the nearest unit, all gifts of
greater value than 100 GBP were deflated into a
‘101⫹’ category. Because of this, and because
the gift values between 1 and 99 GBP were
non-normally distributed, even after transfor-
mation, the dependent variable (gift value) was
split into ordinal categories by terciles.
Throughout the study, alpha level was .05 and
tests were two-tailed. Analysis was conducted
using MLwiN 2.28.
Results
We constructed multilevel cumulative logit
model with a random intercept. The three-
category outcome variable indicated the amount
spent (in GBP) on a gift (“low” ⫽1 – 10;
“medium” ⫽11 – 20; “high” ⱖ21). Table 1
shows the distribution across gift value catego-
ries by relatedness category and network layer
separately for female and male respondents.
The following predictor variables were included
in the initial model: sex (level 2), number of
gifts bought (level 2), relatedness (level 1), net-
work layer (level 1), as well as cross-level in-
teractions between sex and relatedness and be-
tween sex and network layer.
Number of gifts bought was not a significant
predictor and so was dropped from the model.
Checks of the proportional odds assumption
necessitated the inclusion of separate coeffi-
cients for the different spent categories for the
relatedness variable. The small number of
level-1 units (recipients) per level-2 unit (giver)
precluded the inclusion of random slopes for
Table 1
Distribution of Gifts Across Amount Spent Categories by Relatedness Category
and Network Layer for Male Versus Female Gift Givers (Ignoring Nesting
Structure)
Giver gender Recipient category % “low” % “medium” % “high”
Male .5 relative (n⫽111) 19.8 30.6 49.6
Female .5 relative (n⫽187) 18.2 33.7 48.1
Male .25 relative (n⫽47) 46.8 48.9 4.3
Female .25 relative (n⫽78) 50.0 30.8 19.2
Male .125 relative (n⫽10) 80.0 20.0 0.0
Female .125 relative (n⫽35) 37.1 57.1 5.7
Male Friend (n⫽38) 84.2 15.8 0.0
Female Friend (n⫽56) 58.9 14.3 26.8
Male Network layer 1 (n⫽81) 43.2 21.0 35.8
Female Network layer 1 (n⫽105) 28.6 27.6 43.8
Male Network layer 2 (n⫽33) 24.2 24.2 51.5
Female Network layer 2 (n⫽90) 24.4 32.2 43.3
Male Network layer 3 (n⫽92) 44.6 43.5 12.0
Female Network layer 3 (n⫽161) 41.6 35.4 23.0
141DYBLE, VAN LEEUWEN, AND DUNBAR
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

level-1 predictors (as this would have prevented
convergence). The final model is shown in Ta-
ble 2. Note that a higher coefficient represents a
greater probability of being in a lower gift value
category. Results for the random intercept jus-
tify the use of a multilevel model (B⫽1.751,
SE ⫽0.364, p⬍.001).
The model identified a clear relationship be-
tween spending and recipient relatedness (see
Table 2). Givers spent more on gifts for .5
relatives than for .25 (B⫽0.795; SE ⫽0.446;
p⫽.07 for “low” and B⫽1.109; SE ⫽0.489;
p⫽.02 for “medium”), .125 relatives (B⫽
1.883; SE ⫽0.906; p⫽.04 and B⫽3.679;
SE ⫽1.211; p⫽.002), or friends (B⫽4.227;
SE ⫽0.656; p⬍.001 and B⫽⫺3.562; SE ⫽
0.677; p⬍.001). Additional analyses varying
the reference category revealed that signifi-
cantly more money was spent on gifts for .25
and .125 relatives than on friends. However, the
interaction between sex and friend was highly
significant (B⫽⫺2.525; SE ⫽0.727; p⬍
.001). The sign of the coefficient indicates that
women spend more than men on gifts for
friends (relative to .5 relatives). Figure 1 illus-
trates this interaction. With regard to the net-
work layer predictor, givers spent significantly
less on gifts for individuals in network layer 3
as compared with network layer 1 (B⫽1.814;
SE ⫽0.526; p⬍.001). An interaction between
sex and network layer 3 was found, with women
spending more than men on gifts for recipients
in network layer 3 (B⫽⫺1.477; SE ⫽0.627;
p⫽.02). Figure 2 illustrates this interaction.
Discussion
In this study we examined the effects of bi-
ological relatedness, social network position,
and donor sex on self-reported patterns of
Christmas gift giving. As in previous work on
Christmas gift giving (Jonason et al., 2009;
Mysterud, Drevon, & Slagsvold, 2006), gifts of
greater value were bought for kin more closely
related to the gift donor, conforming to the
predictions of kin selection theory. Also in line
with previous work (Mysterud et al., 2006), we
found a sex difference in gift giving, with
women spending significantly more than men
on gifts for friends. Additionally, this study
produced two novel findings. First, we found
that, in line with social network theory, gift
Table 2
Multilevel Cumulative Logit Model for the Effects of Giver Gender, Recipient
Relatedness and Its Interaction With Donor Gender, and Network Layer and Its
Interaction With Donor Gender on Amount of Money Spent on Christmas Gift
Parameter BSEp
Female (ref: male) 0.504 0.505 .32
Network layer (reference: Layer 1)
Layer 2 0.507 0.537 .35
Layer 3 1.814 0.526 ⬍.001
Relatedness – “low” (reference: .5 relatives)
.25 relatives 0.795 0.446 .07
.125 relatives 1.883 0.906 .04
Friends 4.227 0.656 ⬍.001
Relatedness – “medium” (reference: .5 relatives)
.25 relatives 1.109 0.489 .02
.125 relatives 3.679 1.211 .002
Friends 3.562 0.677 ⬍.001
Relatedness ⫻female
.25 relatives ⫻female 0.638 0.541 .24
.125 relatives ⫻female ⫺0.689 1.014 .50
Friend ⫻female ⫺2.525 0.727 ⬍.001
Network layer ⫻female
Layer 2 ⫻female ⫺0.368 0.643 .57
Layer 3 ⫻female ⫺1.477 0.627 .02
Intercept variance 1.751 0.364 ⬍.001
Constant for “low” ⫺2.277 0.427
Constant for “medium” or “low” ⫺0.705 0.408
142 GENDER DIFFERENCES IN CHRISTMAS GIFT-GIVING
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

donors invested more in gifts for recipients in
more proximate social network layers. Second,
we found an interaction between social network
layer and donor gender, with women spending
more than men on gifts for recipients in more
distant network layers.
Taken together, these findings indicate that
men are less willing than women to invest in
their more extended social relationships. If this
reflects a more general sex difference in invest-
ment in social relationships, and an evolved
psychological disposition, why, and in what
context might it have been adaptive? Migration
from one community to another often triggers a
major reconstruction of personal social net-
works, with individuals whom ego was only
distantly associated being promoted to more
proximate network layers. In patrilocal social
systems, women are more likely than men to
migrate and undergo this social network trans-
formation, and therefore might have more to
gain than men in fostering good relations with
more distantly related individuals. If patrilocal
social systems were commonplace during hu-
man evolutionary history, it is plausible that this
could have resulted in a sex-specific psycholog-
ical adaptation that still shapes behavior in evo-
lutionarily novel contexts. Whether patrilocality
was commonplace in human evolutionary his-
tory, however, is unclear. Most contemporary
hunter-gatherers, for example, live in highly
mobile, multilocal groups (Hill et al., 2011).
Some limitations of this study should be not-
ed. First, considerations of reciprocity, fairness,
respect, and hierarchy are also important in
shaping gift-giving decisions—they are not
purely a function of good-will. Reputation, gos-
sip, and even some form of punishment might
face a gift giver who deviates from what is
expected of them. Second, the participants of
this study represented a narrow age range of
young adults. Social network studies have
shown that the composition of personal social
networks change over the lifetime (Saramäki et
al., 2014). Further work could be done in the
same context to see whether the observed sex
differences persist into later life. Finally, the
respondents represent what Henrich, Heine, and
Norenzayan (2010) have called a ‘WEIRD’
population – Western, Educated, Industrialized,
Rich, and Democratic. In this study, this bias is
not just an artifact of the method of data collec-
tion but also of the subject of study itself—the
kind of Christmas gift-giving we report on in
this study is practiced largely in ‘WEIRD’
countries. A sample of different kinds of gift-
giving in other populations would clearly be
desirable.
References
Barrett, D. P., & Hanzi, S. P. (2001). The utility of
grooming in baboon troops. In R. Noë, P. Ham-
merstin, & J. A. R. A. M. van Hooff (Eds.), Eco-
nomics in nature: social dilemmas, mate choice,
and biological markets (pp. 119–145). Cambridge,
UK: Cambridge University Press.
Figure 1. Predicted male versus female spending patterns
on friends in network layer 1. Male respondents shown by
the diagonal bars, females by the dotted bars.
Figure 2. Predicted male versus female spending patterns
on .5 relatives in network layer 3. Male respondents shown
by the diagonal bars, females by the dotted bars.
143DYBLE, VAN LEEUWEN, AND DUNBAR
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Caplow, T. (1982). Christmas gifts and kin networks.
American Sociological Review, 47, 383–392.
Cheal, D. (1988). The gift economy. London, UK:
Routledge.
Colmenares, F., Zaragoza, F., & Hernandez-Lloreda,
M. V. (2002). Grooming and coercion in one-male
units of hamadryas baboons: Market forces or re-
lationship constraints? Behaviour, 139, 1525–
1553. doi:10.1163/15685390260514753
Curry, O., Roberts, S. G., & Dunbar, R. I. (2013).
Altruism in social networks: Evidence for a ‘kin-
ship premium.’ British Journal of Psychology,
104, 283–295. doi:10.1111/j.2044-8295.2012
.02119.x
Henrich, J., Heine, S. J., & Norenzayan, A. (2010).
The weirdest people in the world. Behavioral
and Brain Sciences, 33, 61–83. doi:10.1017/
S0140525X0999152X
Hill, K. R., Walker, R. S., Bozioevic, M., Eder, J.,
Headland, T., Hewlett, B.,...Wood, B. (2011).
Co-residence patterns in hunter-gatherer societies
show unique human social structure. Science, 331,
1286–1289. doi:10.1126/science.1199071
Iredale, W., Van Vugt, M., & Dunbar, R. (2008).
Showing off in humans: Male generosity as a
mating signal. Evolutionary Psychology, 6, 49–61.
Jonason, P. K., Cetrulo, J. F., Madrid, J. M., &
Morrison, C. (2009). Gift-giving as a courtship or
mate-retention tactic? Insights from non-human
models. Evolutionary Psychology, 7, 89–103.
Laroche, M., Saad, G., Kim, C., & Browne, E.
(2000). A cross-cultural study of in-store informa-
tion search strategies for a Christmas gift. Journal
of Business Research, 49, 113–126. doi:10.1016/
S0148-2963(99)00008-9
Mauss, M. (1954). The gifts: Form and functions of
exchange in archaic societies. New York, NY:
Norton.
Mysterud, I., Drevon, T., & Slagsvold, T. (2006). An
Evolutionary Interpretation of Gift-giving Behav-
ior in Modern Norwegian Society. Evolutionary
Psychology, 4, 406–425.
Roberts, S. G. B., Wilson, R., Fedurek, P., & Dunbar,
R. I. M. (2008). Individual differences and per-
sonal social network size and structure. Personal-
ity and Individual Differences, 44, 954–964. doi:
10.1016/j.paid.2007.10.033
Saad, G. (2007). The evolutionary bases of consump-
tion. Mahwah, NJ: Erlbaum.
Saad, G. (2011). The consuming instinct: What juicy
burgers, Ferraris, pornography, and gift giving
reveal about human nature. Amherst, NY: Prome-
theus Books.
Saad, G., & Gill, T. (2003). An evolutionary psychol-
ogy perspective on gift giving among young
adults. Psychology & Marketing, 20, 765–784. doi:
10.1002/mar.10096
Saramäki, J., Leicht, E., López, E., Roberts, S., Reed-
Tsochas, F., & Dunbar, R. I. M. (2014). The per-
sistence of social signatures in human communi-
cation. PNAS Proceedings of the National
Academy of Sciences of the United States of Amer-
ica, 111, 942–947. doi:10.1073/pnas.1308540110
Sherry, J. F. Jr. (1983). Gift giving in anthropological
perspective. Journal of Consumer Research, 10,
157–168. doi:10.1086/208956
Vahed, K. (1998). The function of nuptial feeding in in-
sects: A review of empirical studies. Biological Review,
73, 43–78. doi:10.1017/S0006323197005112
Received December 14, 2013
Revision received August 18, 2014
Accepted August 21, 2014 䡲
144 GENDER DIFFERENCES IN CHRISTMAS GIFT-GIVING
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.