Predicting mothers’ decisions to introduce complementary feeding at 6 months. An investigation using an extended theory of planned behaviour
In Australia and other developed countries there is poor adherence to guidelines recommending the introduction of complementary feeding to infants at 6 months of age. We aimed to investigate, via adopting a theory of planned behaviour framework and incorporating additional normative and demographic influences, mothers' complementary feeding intentions and behaviour. Participants were 375 primiparas who completed an initial questionnaire (infant age 13±3 weeks) that assessed the theory of planned behaviour constructs of attitude, subjective norm, and perceived behavioural control, as well as group norm and additional maternal and infant variables of mothers' age, education level, weight status perception, current maternal feeding practices, and infant birth weight. Approximately, 3 months after completion of the main questionnaire, mothers completed a follow-up questionnaire that assessed the age in months at which the infant was first introduced to solids. The theory of planned behaviour variables of attitude and subjective norm, along with group norm, predicted intentions, with intention, mothers' age (older more likely), and weight status perception (overweight less likely) predicting behaviour. Overall, the results highlight the importance of attitudes, normative influences, and individual characteristics in complementary feeding decision-making which should be considered when designing interventions aimed at improving adherence to current maternal feeding guidelines.
This is the author’s version of a work that was submitted/accepted for pub-
lication in the following source:
Hamilton, Kyra,Daniels , Lynne A,White, Katherine M.,Murray, Nicole, &
Walsh, Anne M. (2011) Predicting mothers’ decisions to introduce comple-
mentary feeding at 6 months : an investigation using an extended theory
of planned behaviour. Appetite,56(3), 674-681 . (In Press)
This ﬁle was downloaded from:
Copyright 2011 Elsevier B.V.
Notice:Changes introduced as a result of publishing processes such as
copy-editing and formatting may not be reﬂected in this document. For a
deﬁnitive version of this work, please refer to the published source:
Running head: mothers’ complementary feeding decision-making
Predicting mothers’ decisions to introduce complementary feeding at 6 months. An
investigation using an extended theory of planned behaviour
Kyra Hamilton1,2, BNursing, BPsych(Hons)
Lynne Daniels2,3, Grad Dip Nut Diet, MSc, PhD
Katherine M. White1,2, BA(Hons), PhD
Nicole Murray2,3, BHsc(Hons)
Anne Walsh2,4, Dip App Sci, BA, Grad Dip HProm, MHSc, PhD
1School of Psychology and Counselling, Queensland University of Technology, Victoria Park
Road, Queensland, 4059, Australia.
2Institute of Health and Biomedical Innovation (IHBI), Queensland University of Technology,
Victoria Park Road, Queensland, 4059, Australia.
3School Public Health, Queensland University of Technology, Victoria Park Road,
Queensland, 4059, Australia.
4School of Nursing and Midwifery, Queensland University of Technology, Brisbane,
Queensland, 4059, Australia.
Corresponding author: Lynne Daniels, Institute of Health and Biomedical Innovation (IHBI),
School of Public Health, Queensland University of Technology, Brisbane, Queensland,
Australia, 4059. Email: firstname.lastname@example.org Phone: +61(07) 3138 6139 Fax: +61(07) 3138
In Australia and other developed countries there is poor adherence to guidelines
recommending the introduction of complementary feeding to infants at 6 months of age.
We aimed to investigate, via adopting a theory of planned behaviour framework and
incorporating additional normative and demographic influences, mothers’ complementary
feeding intentions and behaviour. Participants were 375 primiparas who completed an initial
questionnaire (infant age 13±3weeks) that assessed the theory of planned behaviour constructs
of attitude, subjective norm, and perceived behavioural control, as well as group norm and
additional maternal and infant variables of mothers’ age, education level, weight status
perception, current maternal feeding practices, and infant birth weight. Approximately, 3
months after completion of the main questionnaire, mothers completed a follow-up
questionnaire that assessed the age in months at which the infant was first introduced to
solids. The theory of planned behaviour variables of attitude and subjective norm, along with
group norm, predicted intentions, with intention, mothers’ age (older more likely), and weight
status perception (overweight less likely) predicting behaviour. Overall, the results highlight
the importance of attitudes, normative influences, and individual characteristics in
complementary feeding decision-making which should be considered when designing
interventions aimed at improving adherence to current maternal feeding guidelines.
Key words: complementary feeding, introduction to solids, theory of planned behaviour,
Currently in Australia and other developed countries there is a lack of research
examining the decision-making processes of complementary feeding practices (otherwise
termed introduction of solids). The theory of planned behaviour (TPB; Ajzen, 1991) is a well-
validated behavioural decision-making model that has been used to predict social and health
behaviours (Armitage & Conner, 2001), including maternal breastfeeding practices (e.g.,
McMillan et al., 2008). The aim of the current study was to investigate, using a TPB
framework, mothers’ complementary feeding intentions and behaviour.
A key developmental and nutritional milestone for infants is commencement of
complementary feeding, whereby an infant previously fed only breast milk or formula is
introduced to a wide variety of foods (National Health and Medical Research Council, 2003).
In 2003 the World Health Organisation adopted a ‘global public health recommendation
[that] infants should be exclusively breastfed for the first six months of life to achieve optimal
growth, development and health’. Exclusive breastfeeding is defined as no other fluids
including water or food (World Health Organisation, 2003). Australian guidelines were
modified accordingly to also recommend exclusive breastfeeding for the first 6 months of life
(National Health and Medical Research Council, 2003). Following from this recommendation
is the reciprocal guideline that the introduction of complementary foods (i.e. any solid or
liquid food additional to breast milk or formula; Agostoni et al., 2008), should be delayed
until 6 months of age (National Health and Medical Research Council, 2003). Given that the
introduction of complementary feeding prior to 6 months precludes adherence to the
recommendation to exclusive breastfeeding for 6 months, it is difficult to independently
verify outcomes and, hence, establish evidence separately for either of these
recommendations. Generally, the evidence for the complementary feeding guidelines focuses
on the benefits of exclusive breastfeeding rather than potential independent negative
outcomes of introduction of solids between 4 and 6 months (Arden, 2010). Although the
benefits of exclusive breast feeding are well established, particularly in developing countries,
there are few studies that have examined independent outcomes of delaying introduction of
solids to 6 months in formula fed infants who, by definition, are not exclusively breastfed.
Australian guidelines recommend the introduction of solids ‘at around’ 6 months to
meet the increased nutritional and developmental needs of infants (National Health and
Medical Research Council, 2003). It is argued that earlier solid introduction shows no
benefits and, particularly prior to 4 months, may be associated with negative outcomes such
as inadequate nutrient and energy intake due to displacement of breast milk and formula and
stress on immature gastrointestinal, immune, and renal systems (Arden, 2010; Kaye,
Patterson, Croaker, Norton, & Lewis, 2008; Naylor & Morrow, 2001). Despite these clear
recommendations to the contrary, many mothers introduce solids before their child reaches 6
months of age. A 2003 telephone survey of 1201 children under 5 years in Queensland,
Australia reported that 18% and 67% of infants started complementary foods before the ages
of 4 and 6 months, respectively (Gabriel, Pollard, Suleman, Coyne, & Vidgen, 2005). A
recent representative United States study reported 51% of mothers had introduced solids by 4
months (Grummer-Strawn, Scanlon & Fein, 2008). Similar prevalence rates have been found
in other developed countries (Bolling, Grant, Hamlyn, & Thornton, 2007; Hetzner, Razza, &
Despite the importance of timely introduction of solids and the widespread poor
adherence to the guidelines, studies examining the potentially modifiable behavioural factors
influencing complementary feeding decisions in developed countries are scarce. Most studies
(e.g., Alder et al., 2007; Scott, Binns, Graham, & Oddy, 2009; Wright , Parkinson & Drewett,
2004) have examined factors associated with weaning prior to 4 months of age which was
consistent with the old guidelines. Only a single recent study (N=140 well educated mothers)
from the United Kingdom (UK) has explicitly examined a range of factors important to the
decision to introduce solids at 6 months of age (see Arden, 2010). Even fewer studies have
used an established theoretical framework to explore maternal enablers and barriers to timely
solid introduction (Brophy-Herb, Silk, Horodynski, Mercer, & Olson, 2009) or considered
potential infant and maternal covariates. A systematic review of 33 studies, 7 of which were
from developed countries, identified a range of interrelated social factors such as young
maternal age, lower socioeconomic status, lower educational achievement, ethnicity, and
formula feeding that are associated with early complementary feeding (see Lanigan, Bishop,
Kimber, & Morgan, 2001). While these largely demographic maternal and infant
characteristics are important, the majority is not readily modifiable and do not explain the
complexity of social and psychological influences that underpin decisions about
complementary feeding. Given the reciprocal relationship between the breastfeeding and
complementary feeding guidelines, understanding these factors is also important in improving
duration of exclusive breastfeeding. Understanding psychosocial predictors will allow us to
assess if current messages accurately target the beliefs and behaviours that contribute to poor
adherence to both the breastfeeding and complementary feeding guidelines.
Theory of planned behaviour
The theory of planned behaviour (TPB; Ajzen, 1991) is a well-validated behavioural
decision-making model, widely employed to examine the psychosocial influences on
behaviour. The TPB proposes the most proximal determinant of behavioural outcomes is
intention to perform a given behaviour which is, in turn, predicted by three belief-based
constructs: attitudes, subjective norms, and perceived behavioural control (Ajzen, 1991).
Attitudes are the positive or negative evaluations by an individual about the consequences of
performing a particular behaviour. Subjective norms refer to the perceived pressure from
important others to perform or not to perform an action. Perceived behavioural control, which
is similar to the concept of self-efficacy, refers to one’s perceived ease of performing a given
behaviour and is also proposed to influence behaviour directly. The TPB has been used to
examine a wide range of behaviours, including nutritional and dietary practices (e.g.,
Blanchard et al., 2009; Verbeke & Vackier, 2005). A meta-analysis (Armitage & Conner,
2001) found that the TPB accounted for an average of 39% of the variance in people’s
intentions and 27% of the variance in behaviour.
Theory of planned behaviour and the prediction of maternal feeding behaviours
A small number of studies have applied the TPB to examine maternal feeding
practices, most of which are related to breastfeeding behaviours. McMillan et al. (2008)
investigated breastfeeding uptake in primiparas (N=248) experiencing material hardship.
Consistent with the specifications of the TPB, attitudes, subjective norm, and perceived
behavioural control were all significant predictors of mothers’ intentions to breastfeed
(explaining 56% of the variance after controlling for age, ethnicity, education, and
deprivation), with intention and perceived behavioural control predicting breastfeeding
behaviour 6-weeks later (explaining 44% of the variance after controlling for age, ethnicity,
education, and deprivation). Other researchers (e.g., Khoury, Moazzem, Jarjoura, Carothers,
& Hinton, 2005; Swanson & Power, 2005) have found attitudes and subjective norms, but not
perceived behavioural control, to predict breastfeeding initiation and continuation. Given the
success of the TPB model in predicting other maternal feeding behaviours, it is plausible that
this framework would be useful in examining complementary feeding practices.
One of the few studies to use TPB to explore complementary feeding behaviour is
from Horodynski et al. (2007) who used the TPB framework in a qualitative thematic analysis
from six focus groups (N= 23) with low income mothers in the United States. The aim was to
examine knowledge and attitudes regarding introduction of solids in the context of the
recommendation of introduction no earlier than 4-6 months. Mothers knew and approved of
the recommendation but infant factors such as sleep patterns and satiety weakened their
intention to delay introduction. They also identified subjective norms based on social pressure
from families, avoiding negative effects of early solid introduction (rather than positive
outcomes of later introduction), and low perceived behavioural control (e.g., diagnosis of acid
reflux) as important influences on complementary feeding behaviour. However, these
conclusions were based on qualitative rather than quantitative analyses.
Despite the success of the TPB in predicting maternal breastfeeding behaviours, as is
the case for most studies using the TPB as a predictive model, there still remains a proportion
of unaccounted variance. Ajzen (1991) supports the inclusion of additional predictors to the
model to improve its prediction of people’s intentions and/or behaviour. However, there
should be a strong theoretical justification for inclusion of additional predictors and they
should capture a significant portion of unique variance in intentions or behaviour.
Increasingly, researchers have recognized the importance of normative influences from
relevant referent social groups and have included an assessment of group norms in the model
(see e.g., Hamilton & White, 2008; White, Smith, Terry, Greenslade, & McKimmie, 2009),
including in studies investigating food choice behaviours (e.g., Carrus, Nenci, & Caddeo,
2009; Louis, Davies, Smith, & Terry, 2007). Given the popularity of both formal and informal
mothers’ support groups (Campbell et al., 2008), social influences and contexts might be
especially salient for mothers of young children. Several studies have identified family and
friends as a key influence on complementary feeding practices (see Alder et al., 2007;
Crocetti, Dudas, & Krugman, 2004; Horodynski et al., 2007; Olson, Horodynski, Brophy-
Herb, & Iwanski, 2008). Thus, the current study included an assessment of group norms
within the TPB.
Group norms and social identity influence
The subjective norm construct within the TPB reflects injunctive norms as the focus is
on people’s perceptions of social pressure from important others to perform the behaviour
(Ajzen, 1991). Group norms refer to the explicit or implicit prescriptions regarding one’s
appropriate attitudes and behaviours as a member of a specific reference group in a given
context (White, Hogg & Terry, 2002). Thus, where subjective norm infers that perceptions of
social approval from important others influence people’s intentions, group norm infers that
normative support from a relevant group (i.e., other mothers) for performing the behaviour
(e.g., perceptions about whether other group members perform the behaviour themselves;
similar to the concept of the descriptive norm component of the TPB, see Rivis & Sheeran,
2003) influence people’s intentions.
The influence of social identity on the intention-behaviour relationship can be
explained using a social identity (Hogg & Abrams, 1988) and self-categorization theory
(Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) perspective. According to these theories,
when social identity is salient, the individual constructs context-specific group norms based
on shared intra-group information and assimilate themselves to these group norms (Turner et
al., 1987). Behavioural performance, therefore, is more likely to occur when there is support
from a relevant ingroup endorsing the behaviour as a good thing to do and performance of the
behaviour than without such ingroup support (e.g., Terry & Hogg, 1996). Accordingly, group
norms influence behavioural performance as the individual, based on their observations of
group members, seeks to act in a manner similar with their ingroup, therefore achieving
categorization as a group member (Hogg & Abrams, 1988; Turner et al., 1987).
More recent evidence suggests that group norms predict behavioural intentions
irrespective of level of identification (e.g., Johnson & White, 2003). The normative influences
of family and friends have been identified by several studies as a key determinant of
complementary feeding practices (Alder et al., 2007; Crocetti et al., 2004; Horodynski et al.,
2007; Olson et al., 2008). Hence, in the current study, the influence of the perceived action of
an important referent group (i.e., most mothers I know) was specifically examined.
The present study
This study aimed to use the theoretical framework of the TPB, incorporating
additional demographic factors and group norms, to determine predictors of mothers’
intentions and behaviours related to introduction of solids at 6 months of age. The target
behaviour (introducing solids for the first time at 6 months of age) was chosen based on
current national and international guidelines (National Health and Medical Research Council,
2003; World Health Organisation, 2003).
From a TPB perspective, it was hypothesised that attitude, subjective norm, and
perceived behavioural control would predict mothers’ intention to introduce complementary
feeding at 6 months (Hypothesis 1), and intention and perceived behavioural control would
predict the actual introduction of solids at 6 months (Hypothesis 2). In relation to the
normative influence variable, it was expected that group norms would be identified as an
additional predictor of mothers’ intentions to introduce complementary feeding at 6 months
(Hypothesis 3). Furthermore, in an exploratory manner, maternal demographic variables of
age, pre-pregnancy weight status, education level, and current maternal feeding practices, as
well as the infant characteristic of birth weight, were examined to determine if they
contributed to the prediction of intention and behaviour related to the introduction of solids at
6 months (Hypothesis 4).
Participants and procedure
Participants were recruited in 2008 from a consecutive sample of first-time mothers,
aged 18 years or older, who delivered healthy term infants in seven hospitals in the capital
cities (Brisbane and Adelaide) of two Australian states. Eligible mothers were approached
postnatally and gave consent to be contacted once their infants were aged 4-7 months
regarding enrolment in NOURISH, an early feeding randomized controlled trial (Daniels et
al., 2009). At the first postnatal contact they provided brief demographic data and also
indicated if they were willing to be contacted regarding other related studies. Mothers
consenting to the latter formed the sampling frame for the study described here and were
contacted separately when their infant was aged 2 months.
The study used a prospective design with two waves of data collection. The main
questionnaire (infant aged approximately 3 months) assessed the standard TPB predictors
(i.e., attitude, subjective norm, perceived behavioural control, and intentions), along with
group norm, feeding mode, and the additional demographic measures. The follow-up
questionnaire (infant aged approximately 7 months) assessed the age (in months) at which
mothers first introduced solids.
A letter of invitation explaining the study, the main questionnaire, and a reply paid
envelope were mailed to eligible participants. The invitation letter advised mothers that a
follow-up questionnaire would be sent in 3 to 4 months. Mothers who did not wish to
participate were asked to return the blank questionnaire. A reminder letter was sent at two
weeks with a second reminder letter and replacement questionnaire sent at three weeks.
For the first recruitment cohort of NOURISH , 85% (N=1932) of eligible mothers
were approached on the postnatal wards, 74% (N=1422) of whom provided contact details.
Of those who declined to provide contact details, 44% (N=510) agreed to provide brief
demographic data. These data indicated those agreeing to later contact were more likely to
have a university education (41% vs. 22%). There were few differences in other key
demographic variables: birth weight 3.48 vs. 3.46 kg; age 27 vs. 26 years, and 17% in both
groups self-reported themselves as overweight prior to their pregnancy. Of the participants
consenting to be re-contacted for enrolment in the NOURISH trial, 75% (N= 1062) also
provided consent (either active or passive) to be contacted regarding other related studies. Of
these 379 were excluded as their infant turned 3 months prior to finalisation of the
questionnaire and the mail out commencing in June 2008. Thus, the main questionnaire was
sent to the remaining 683 mothers (43% in Adelaide) when their infant was approximately 3
months old. It was estimated that a sample size of N=370 was needed to provide 95%
confidence that results in the sample will be the same as the population within a 5% error
range, assuming a 40/60 split on any variable (de Vaus, 2002) and we assumed a 55%
response rate. In the TPB study reported here, 375 (55% response rate) first-time mothers
completed the main questionnaire.
Fifty-four percent of those who returned the main questionnaire subsequently enrolled
in the NOURISH trial (Daniels et al., 2009). Comparison on age of introduction of solids and
key covariates revealed no differences between those in the larger study and those who
consented for the TPB study. Approval for recruitment to the larger study was obtained from
the relevant hospital human research ethics committees with the ethics committees for both
Queensland University of Technology and Flinders University approving the TPB study
The target behaviour of introducing solids was defined as, “any foods or drinks given to your
baby in addition to breast milk, formula, or water”, and introducing solids for the first time at
6 months of age was the defined timeframe. Examples of solids were presented and included:
homemade foods that are easy to eat such as mashed fruits or vegetables; commercial baby
foods such as rice cereal, tins or jars of baby foods; and drinks like cow’s milk or juice. To
maximise congruence between the prediction and criterion variables, the TPB variables were
measured at the same level of specificity in terms of action, target, and time (Ajzen &
Fishbein, 1970). In this study we only examined attitudes and beliefs that were directly and
uniquely related to the specific target behaviour and defined time frame (i.e., solid
introduction at age 6 months). The TPB items were constructed in line with recommendations
(Ajzen, 1991) and were each scored on a 7-point Likert scale, except for attitude, which was
scored on 7-point semantic-differential scales.
Intention. Three items assessed the strength of intention to perform the target
behaviour (e.g., “I intend to introduce solids when my baby is 6 months old”, scored strongly
disagree  to strongly agree ).
Attitude. Attitude towards introducing solids at 6 months was assessed by three, 7-
point semantic differential scales (e.g., “For me, introducing solids (for the first time) when
my baby is 6 months of age would be… unimportant  to important ).
Subjective norm. Subjective norm was assessed by two items (e.g., “Most people
who are important to me think that I should introduce solids when my baby is 6 months of
age”, scored strongly disagree  to strongly agree ).
Perceived behavioural control. Perceived behavioural control was measured by three
items reflecting the mother’s sense of control about performing the target behaviour (e.g., “I
have complete control over whether or not I introduce solids when my baby is 6 months of
age”, scored strongly disagree  to strongly agree ).
Group norm. Group norm was measured by one item, (e.g., “Most mothers I know
introduce solids when their babies are 6 months of age”, scored strongly disagree  to
strongly agree ). This reference group was based on outcomes of a small elicitation study
Demographics variables. A number of demographic details were collected at the
first postnatal contact including: (i) mothers’ age at the birth of the infant in years; (ii)
mother’s self-reported weight status, (“How would you describe your weight status before
pregnancy”), coded as 1 = not overweight and 2 = overweight; and (iii) education levels coded
as 1 = non-university and 2 = university.
Current feeding practices at approximate age of 3 months were assessed in the main
questionnaire using 1 of 6 responses (exclusive breastfeeding , mainly breastfeeding with
other drinks occasionally , breastfeeding and solids , formula feeding , formula
feeding and solids , breastfeeding and formula feeding ). Responses from 1 to 3,
indicating the individual is primarily breastfeeding (that is not using formula), were coded as
1 = breastfeeding and mothers who responded 4 to 6, indicating they were using formula,
were coded as 2 = formula feeding. Finally, infants’ birth weight (kilograms) was recorded
from hospital records.
Reported behaviour. Approximately 3 months after completion of the main
questionnaire, mothers were sent a follow-up questionnaire that included the question “At
what age was your baby first given solid foods? ____ months”.
Three hundred and seventy-five (55% response rate) first-time mothers, mean age of
29.2 (±5.5) years at the birth of their baby, completed the main questionnaire. One hundred
and sixty-nine of the mothers (45%) had university degree qualifications with 61 (16%)
mothers perceiving themselves as overweight prior to pregnancy. Infants had a mean birth
weight of 3.47(±0.50) kg. At completion of the main questionnaire, babies were aged 13(±3)
weeks and 32 (9%) had been given solid foods. Two hundred and nine (56%) of the mothers
were currently breastfeeding (i.e., exclusive breastfeeding, n = 190, 51%; and breastfeeding
with small amounts of other drinks or solids, n = 19, 5%). One hundred and sixty-five (44%)
of the mothers were using formula (i.e., exclusive formula, n = 93, 25%; formula and solids, n
= 13, 3%; and breastfeeding and formula, n = 59, 15.7%). Of those who returned the main
questionnaire, 69% (257) returned the follow-up questionnaire which provided the age of
solid introduction data. The mean age of the babies at this time point was 31(±2) weeks.
Compared to participants who failed to return the follow up questionnaire, mothers who
completed both questionnaires were more likely to have a university education (52% vs. 33%,
p=0.001) and be older (29.9 years vs. 28.0 years, p=0.002), and were less likely to report
themselves as overweight prior to pregnancy (14% vs. 22%, p=0.055) and to be using formula
at the time they responded to the first questionnaire (43% vs. 61%, p=0.001). There was no
significant difference in infant birth weight. Only those providing complete data for all the
demographic and TPB construct and criterion variables were included in the regression
The means, standard deviations, correlations, and reliabilities of the variables are
reported in Table 1. Low to moderate correlations were found amongst the TPB predictors
(attitude, subjective norm, and perceived behavioural control), all of which correlated with
intention. Attitude and subjective norm, but not perceived behavioural control, were
correlated with behaviour, with intention emerging as the strongest behavioural correlate.
Group norm showed low to moderate correlations with attitude, perceived behavioural
control, and the TPB criterion variables of intention and behaviour, and a strong association
with subjective norm. Similarly, low to moderate correlations were found between the
demographic variables of maternal age and current feeding practices and intention, and
maternal pre-pregnancy self-reported weight status, education, age, and current feeding
practices and behaviour. The average age at which infants were first introduced to solids was
5 (±0.84) months. All of the measures were reliable, with a slightly lower reliability
evidenced for perceived behavioural control.
Regression analysis predicting intentions
A hierarchical multiple regression analysis examined the proposed predictors of
mothers’ intentions to introduce solids at 6 months. The demographic factors were entered at
step 1, with the standard TPB variables being entered at step 2 and group norm at step 3. The
step 1 variables accounted for 4% of the variance in intentions, F(5, 352) = 2.75, p =.019,
with mothers’ age (older more likely to intend) and current feeding practices (breastfeeding
more likely to intend) revealed as significant. The step 2 variables significantly accounted for
an additional 61% of the variance in intentions, F(3, 349) = 15.65, p < .001. The TPB
variables of attitudes and subjective norms, but not perceived behavioural control, were found
to be significantly related to intentions. Inclusion of the TPB variables at step 2 revealed the
demographic factors of mothers’ age and current feeding practices as no longer significant
predictors of intention. Group norm entered at step 3 accounted for a further 0.5% of the
variance in intentions, F(1, 348) = 4.90, p = .027. When all the variables were entered into the
equation at step 3, the significant predictors explaining 65% of the variance in intentions were
attitude, subjective norms, and group norms (see Table 2).
Regression analysis predicting behaviour
An additional hierarchical multiple regression analysis examined the proposed
predictors of introducing solids at 6 months. The demographic factors were entered at step 1,
with intention and perceived behavioural control entered at step 2, and attitude, subjective
norm, and group norm at step 3. As shown in Table 3, step 1 explained 15% of the variance,
F(5, 230) = 8.15, p <.001, with maternal age (older more likely to introduce solids later),
current feeding practices (formula feeders more likely to introduce solids earlier), and
mothers’ perceived weight status (overweight more likely to introduce solids earlier) reported
as significant. The addition of step 2 accounted for a further 17% of the variance in behaviour,
F(2, 228) = 28.23, p <.001. Mothers’ age and weight status perception, but not current feeding
practices, remained significant predictors of behaviour, with intention also found to be
associated with mothers’ introduction of solids at 6 months. Attitude, subjective norm, and
group norm entered at step 3 did not significantly account for an additional amount of the
variance in behaviour, F(3, 225) = 1.51, p = .212. In the overall model, mothers’ age,
perceived weight status, and intentions were the significant predictors and accounted for 33%
of the variance of later solid introduction.
This is one of the first studies to use an established theoretical framework, the TPB, to
inform a quantitative examination of psycho-social and demographic factors associated with
mothers’ adherence to the current guidelines to introduce solids to their infant at 6 months of
age. The results provide partial support for the TPB in that attitude and subjective norms
predicted intentions with intentions, in turn, predicting mothers’ actual behaviour of
introducing solids at 6 months. Contrary to TPB, perceived behavioural control did not
predict either intention or behaviour. As expected, the additional normative construct of group
norm was associated with intention. In the final models there was no association between
intention and any of the demographic variables, including current feeding and education, and
only intention, maternal self-reported weight status, and age predicted behaviour. Overall, our
results indicate that older mothers with self-reported normal weight status and strong
intentions at 3 months were more likely to comply with guidelines to introduce solids at 6
months of age.
In partial support of Hypothesis 1, and consistent with the TPB in general (Armitage
& Conner, 2001) and its specific application to investigating breastfeeding behaviours (e.g.,
McMillan et al., 2008), attitude and subjective norm predicted mothers’ intentions to
introduce solids at 6 months. These findings suggest that mothers who more favorably
evaluate and perceive pressure from important others to introduce solids at 6 months will have
stronger intentions to do so. In the current study, subjective norm was found to be a stronger
predictor than attitudes which is contrary to many TPB studies that find subjective norms to
be the weakest predictor of people’s intentions (Armitage & Conner, 2001). Thus, our results
highlight the important contribution of positive attitudes and especially social approval to
mothers’ intentions to introduce solids at 6 months.
The importance of perceived benefits of solid introduction at 6 months (i.e., attitudes),
although not as strong as the effect of perceived social pressures (i.e., subjective norms), in
informing mothers’ decisions about complementary feeding is supported somewhat by prior
research. In their qualitative study, Horodynski et al. (2007) concluded that avoiding potential
negative effects of early introduction was a more important predictor than the positive
outcomes of later introduction. In Horodynski et al.’s study, the definition of early
introduction was prior to 4 months and the findings may realistically reflect the relative
strength of the evidence for detrimental outcomes related to introduction prior to 4 rather than
6 months. Arden (2010) administered an electronic questionnaire to 140 well-educated UK
mothers (67% with a university degree) to assess the importance (7 points; ‘not at all’ to
‘extremely’ important) of 23 items in making the decision to introduce solid foods to their
baby at 6 months. The target children had a mean age 18.5 (6-36) months and hence
participants were providing a retrospective recall of their decision making. Regression
analysis identified that the perceived importance of the current WHO guidelines to introduce
solids at 6 months was by far the strongest predictor of age at which solids were first
introduced in this sample of very well educated mothers. Together, these data imply that
positive evaluations about the guidelines are very important in establishing intention to
introduce solids at age 6 months which, in turn, was the only TPB construct that predicted
For the additional normative influence measure included in the present study, group
norm (along with subjective norm) emerged as a significant predictor of mothers’ intentions
to introduce solids at 6 months, supporting Hypothesis 3. The significant influence of group
norms suggests that mothers’ intentions to introduce complementary feeding at 6 months are
stronger if they perceive that other mothers they know perform the behaviour. Overall, these
results highlight the impact of the relationship between normative influences and behavioural
intentions and emphasize the importance of direct social pressures from important referents
(subjective norms) and to some extent groups and other mothers (group norms), in providing
normative information that mothers consider when deciding how to behave in this context.
These findings concur with research that suggest normative influences are important
determinants of introducing solids early (Crocetti et al., 2004; Horodynski et al., 2007).
In their TPB qualitative study, Horodynski et al. (2007) also identified social pressure
from family members as an important influence on complementary feeding behaviour. Wright
et al. (2004) found that mothers in their Millennium Baby Study returning questionnaire data
on weaning (N=707, 77%) who started solids early (<13 weeks) were more likely than those
starting later (> 17 weeks) to cite the influence of family and friends as a reason for starting
solids (40% vs 25%, p= χ2 0.024). Furthermore, Olson et al. (2008) found that health
professionals (N=36) participating in focus groups (N=5) reported that female relatives, in
particular mothers, were the strongest influence infant feeding practices. A recent pilot
intervention effectively incorporated explicit content and role playing designed to enhance
mothers’ self efficacy in defining their feeding intentions and discussing these with, and
dealing with conflicting advice from, social groups (Brophy-Herb et al., 2009). Participants
were 31 low income mothers (16 African Americans) and the intervention was delivered
during six home visits over 6 weeks in the context of other support programs for
disadvantaged mothers. In Arden’s (2010) UK study, support from family members emerged
as a theme in the qualitative analysis, but was not confirmed as predictor of introduction of
solids at 6 months in the quantitative analysis. In our study, and as expected within the TPB
framework (Ajzen, 1991), social norms (i.e., subjective norm and group norm) predicted
intentions. Thus, it may be important to provide anticipatory guidance for both mothers and
their significant others as early as possible to define and support mothers’ intentions to
introduce solids at 6 months and hence adherence to the current guidelines.
Consistent with Hypothesis 2, strong intentions to introduce solids at 6 months
predicted self-reported behavioural performance. Contrary to Hypothesis 1 and 2, perceived
behavioural control did not emerge as a predictor of intention or behaviour. This finding is
inconsistent with results from a study of breastfeeding behaviour in disadvantaged primiparas
where support for the predictive ability of perceived behavioural control for both intention
and behaviour performance was found (McMillan et al., 2008). Although our slightly lower
reliability of the perceived behavioural control scale suggests the need for cautious
interpretation, others investigating breastfeeding practices have found similar results to our
study (e.g., Swanson & Power, 2005).
According to Ajzen (1991), the strength of perceived behavioural control in
determining behaviour is dependent on perceptions of control matching actual control. Thus
our results, which show no association between perceived behavioural control and behaviour
or intention, suggest that mothers might not have been accurate in estimating their control
over their intentions and behavioural performance (see Sheeran, Trafimow, & Armitage,
2003). A plausible alternate interpretation is that mothers may be encouraged to respond to
their infants needs and, hence, perceived control lies more with the infant than themselves.
Several studies have identified the influence of signs of readiness from the infant, including
weight and hunger, as an important influence in the decision to introduce solids (Arden, 2010;
Olsen et al., 2008; Wright et al., 2004). Arden (2010) discusses the contradiction implicit in
the need to be responsive to their infants needs in order to meet recommendations to demand
breast or formula feed and yet potentially ignore signs of readiness in order to adhere to
introduction of solids guidelines. Arden (2010) and others (Olson et al., 2008) further
highlight the lack of evidence and clarity regarding which are true signs of readiness. Beliefs
regarding infant readiness were considered to potentially influence introduction of solids at
any age and, thus, were not uniquely relevant to the specific target behaviour in our study.
Hence, items addressing readiness beliefs were not included in our TPB constructs. Future
research should examine the influence of beliefs about infant readiness and perceptions of
infant needs on intentions and its role in the introduction of complementary feeding at 6
months of age.
In consideration of previous literature supporting a role of demographic factors in
predicting early complementary feeding (Lanigan et al., 2001), in an exploratory manner, we
investigated the impact of infant and maternal demographic factors, within the TPB, on
intention and behaviour to introduce solids at 6 months (Hypothesis 4). The demographic
variables had minimal predictive value for intention but maternal age, pre-pregnancy weight
status, and current feeding mode accounted for 15% of the variance in reported behaviour.
However, after the inclusion of the social-cognitive factors, only maternal age and self-
reported weight status perception remained significant predictors of the actual introduction of
solids at 6 months. Older mothers who did not perceive themselves to be overweight were
more likely to introduce solids at 6 months. Although gestational obesity has been associated
with reduced initiation and duration of breastfeeding (Amir & Donath, 2007; Rasmussen &
Kjolhede, 2007), we are unaware of other studies that have shown maternal weight status to
be an independent predictor of early solid introduction. A number of studies, however, have
identified young maternal age as an important risk factor for early solid introduction (Bolling
et al., 2007; Coleman et al., 2009; Scott et al., 2009; Wright et al., 2004).
In contrast to our study, where current feeding mode (at 3 months of age) was not an
independent predictor of introduction of solids at 6 months, others have shown not
breastfeeding at 4 months (Crocetti et al., 2004; Wright et al., 2004) and not exclusively
breastfeeding at 4 weeks (Scott et al., 2009) predicted early (prior to 4 months) solid
introduction. These differences may relate to the comparatively high rate of breastfeeding and
low rate (9%) of early solid introduction (prior to 3-4 months) in our sample and/or the target
age of solid introduction being 6 rather than 4 months. The lack of predictive association of
educational achievement (tertiary vs less than tertiary) with introduction of complementary
feeding later than 4 months is consistent with two recent Australian studies (Gabriel et al.,
2005; Scott et al., 2009). In contrast, a recent five-country European study (Schiess et al.,
2010) reported that low maternal education was positively associated with introduction of
solids by 3 and 4 completed months of age. These data are from a randomised controlled trial
evaluating protein content of infant formula which included twice the number of formula-fed
infants compared to breastfed infants. Given the strong and well established relationships
between maternal education, socioeconomic status, and formula use and early solid
introduction (Alder et al., 2007; Bolling et al., 2007; Scott et al., 2009), the relatively high
prevalence of breastfeeding in our sample, the target age of solid introduction being 6 months,
and the reporting of university vs. no university education as opposed to reporting on years of
education may explain the disparity with our results.
The research has a number of strengths including the prospective examination of a
unique and specifically defined behaviour within a TPB framework, the examination of the
impact of a range of covariates, and a relatively large sample size with an acceptable response
rate. The current study also has a number of limitations. The use of self-report data may
facilitate socially desirable responses. The sample was predominately Caucasian and, hence,
the relevance of the findings to mothers from other cultural backgrounds is uncertain. In
addition, the sample consisted of only primiparous women. Although previous research
suggests that feeding behaviours applied to the first child are strongly predictive of feeding
choices with subsequent children (Bolling et al., 2007; De Vanzo, Starbird, & Leibowitz,
1990), future research could examine the efficacy of the TPB for women having their second
and subsequent children. Furthermore, conclusions about the results for group norm should be
interpreted with caution given assessment used a 1-item self-report measure which was highly
correlated with subjective norm. Future research should employ a multi-item scale of group
norm to confirm the current findings. Finally, although the models in the current study
explained a substantial amount of variance in both intentions and behaviour, two thirds of the
variance in the target behaviour, age of solid introduction, remains unexplained. Future
research needs to consider other variables which might predict the introduction of
complementary feeding at 6 months. For example, future studies should examine how social
support factors (e.g., advice from family members, friends, or healthcare professionals),
caregiver knowledge (see Crocetti et al., 2004; Horodynski et al., 2007), and concepts of baby
readiness (Arden, 2010), within a TPB framework, might influence the introduction of solids
at 6 months of age and, hence, adherence to the guidelines.
Despite the limitations of this study, the results highlight some important
considerations when designing interventions aimed at improving adherence to current infant
feeding guidelines. Specifically, the results point to the importance of attitudes, normative
influences, and individual characteristics in complementary feeding decision-making. In
particular, our results indicate that believing that introducing solids at age 6 months is
important and that this belief is shared by significant others in the mother’s life are the
principal modifiable predictors of intention to complementary feeding introduction. These
findings imply that strategies that focus on improving the understanding of the rationale
underlying the current guidelines in mothers and their families are important. Given that this
intention at infant age 3 months is the only modifiable factor that predicts actual behaviour, it
is important that mothers and significant family members are encouraged as early as possible
to consider and plan the timing of solid introduction. Our data suggest that anticipatory
guidance in both antenatal and postnatal settings should specifically promote timely
introduction of solids in conjunction with initiation and maintenance of breastfeeding. This
strategy is particularly relevant given the interdependence of guidelines concerning the
duration of exclusive breastfeeding and timing of introduction of solids. Mothers choosing to
formula feed also require early support to foster the intention to introduce solids at 6 months
of age. Furthermore, it may be useful to explicitly target other members of the mother’s
network, including other mothers they know, to help frame the introduction of solids at 6
months as a normative behaviour. The results also suggest younger mothers and those who
are overweight should particularly be targeted for this advice. These strategies are likely to
support mothers in their intention to introduce solids at 6 months.
In conclusion, the current study, via adopting a TPB framework, provides a basis for
understanding complementary feeding practices and important applied information that can be
used in developing strategies that inform mothers and their families on the benefits of and
support adherence to current guidelines recommending introducing solids no earlier than at 6
months of age. Specifically, the findings suggest that attention to establishing positive
attitudes toward introducing solids at 6 months and considering the normative influences of
others, such as the approval from important referents and the behavioural practices of other
mothers, may assist in improving mothers’ decision-making to adhere to recommended
national guidelines, thus maximising the benefits to the health and well-being of the infant.
The authors would like to thank Sierra Van Wyk for her research assistance on this project
and extend appreciation to all the mothers who volunteered to participate in this study. This
recruitment of participants in this study was funded by the National Health and Medical
Research Council. Nicole Murray was supported by doctoral scholarship from Queensland
University of Technology.
Agostoni, C., Decsi, T., Fewtrell, M., Goulet, O., Kolacek, S., Koletzko, B., et al.
(2008). Complementary feeding: a commentary by the ESPGHAN Committee on
Nutrition. Journal of Pediatric Gastroenterology and Nutrition, 46, 99-110.
Ajzen, I. (1991). The theory of planned behaviour. Organizational Behavior and
Human Decision Processes, 50, 179-211.
Ajzen, I., & Fishbein, M. (1970). The prediction of behavior from attitudinal and normative
variables. Journal of Experimental Social Psychology, 6, 466-487.
Alder, E. M., Williams, F. L. R., Anderson, A. S., Forsyth, S., Florey, C.V., & van der Velde,
P. (2007). What influences the timing of the introduction of solid food to infants?
British Journal of Nutrition, 92, 527-531.
Amir, L. H. & Donath, S. (2007). A systematic review of maternal obesity and breastfeeding
intention, initiation and duration [Electronic Version]. BMC Pregnancy and
Childbirth. 7, 9 Retrieved from http://www.biomedcentral.com/1471-2393/7/9
Arden, M. (2010). Conflicting influences on UK mothers’ decisions to introduce solid
foods to their infants. Maternal & Child Nutrition, 6, 159-173.
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A
meta-analytic review. British Journal of Social Psychology, 40, 471-499.
Blanchard, C. M., Kupperman, J., Sparling, P. B., Nehl, E., Rhodes, R. E., Courneya,
K. S., & Baker, F. (2009). Do ethnicity and gender matter when using the theory of
planned behaviour to understand fruit and vegetable consumption. Appetite, 52, 15-
Bolling, K., Grant, C., Hamlyn, B., & Thornton, A. (2007). Infant feeding survey 2005.
A survey conducted on behalf of the information centre for health and social care and
the UK Health Departments by BMRB Social Research. London: The Information
Centre for Health and Social Care.
Brophy-Herb, H., Silk, K., Horodynski, M., Mercer, L., & Olson, B. (2009). Key theoretical
frameworks for intervention: understanding and promoting behavior change in parent–
infant feeding choices in a low-income population. The Journal of Primary Prevention,
Campbell, K., Hesketh, K., Crawford, D., Salmon, J., Ball, K., & McCallum, Z. (2008). The
Infant Feeding Activity and Nutrition Trial (INFANT) an early intervention to prevent
childhood obesity: cluster-randomised controlled trial [Electronic Version]. BMC
Public Health, 8, 103 Retrieved from http://www.biomedcentral.com/1471-
Carrus, G., Nenci, A. M., & Caddeo, P. (2009). The role of ethnic identity and perceived
ethnic norms in the purchase of ethnical food products. Appetite, 52, 65-71.
Coleman, B. L., Gutmanis, I., Larsen, L. L., Leffley, A. C., & McKillop, J. M. (2009).
Introduction of Solid Foods. Do mothers follow recommendations? Canadian Journal
of Dietetic Practice and Research, 70, 135-140.
Crocetti, M., Dudas, R., & Krugman, S. (2004). Parental beliefs and practices regarding early
introduction of solid foods to their children. Clinical Pediatrics, 43, 541-547.
Daniels, L., Magarey, A., Battistutta, D., Nicholson, J. M., Farrell, A., Davidson, G., &
Cleghorn, G. (2009). The NOURISH randomised control trial: positive feeding
practices and food preferences in early childhood - a primary prevention program for
childhood obesity [Electronic Version]. BMC Public Health, 9, 387 Retrieved from
De Vanzo, J., Starbird, E., & Leibowitz, A. (1990). Do women’s breastfeeding
experiences with their first borns affect whether they breastfeed their subsequent
children? Social Biology, 37, 223-232.
de Vaus, D. A. (2002). Surveys in social research. Malaysia: Allen & Unwin.
Gabriel, R., Pollard, G., Suleman, G., Coyne, T., & Vidgen, H. (2005). Infant and child
nutrition in Queensland 2003. Brisbane: Queensland Health.
Grummer-Strawn, L. Scanlon, K. & Fein, S. (2008) Infant feeding and feeding transitions
during the first year of life. Pediatrics, 122, S36-S42.
Hetzner, N. M. P., Razza, R. A., & Brooks-Gunn, J. (2009). Associations among
feeding behaviors during infancy and child illness at two years. Maternal and
Child Health Journal, 13, 795-805.
Hamilton, K., & White, K. M. (2008). Extending the theory of planned behavior: the role of
self and social influences in predicting adolescent regular moderate-to-vigorous
physical activity. Journal of Sport & Exercise Psychology, 30, 56-74.
Hogg, M. A., & Abrams, D. (1988). Social identification: a social psychology of intergroup
relations and group processes. London: Routledge.
Horodynski, M., Olson, B., Arndt, M. J., Brophy-Herb, H., Shirer, K., & Shemanski, R.
(2007). Low-income mothers' decisions regarding when and why to introduce solid
foods to their infants: influencing factors. Journal of Community Health Nursing, 24,
Johnston, K. L., & White, K. M. (2003). Binge-drinking: a test of the role of group norms in
the theory of planned behaviour. Psychology and Health, 18, 63-77.
Kaye, J., Patterson, P., Croaker, S., Norton, L., & Lewis, F. (2008). A healthy start in
life. A nutrition manual for health professionals. Brisbane: Queensland Health.
Khoury, A. J., Moazzem, S. W., Jarjoura, C. M., Carothers, C., & Hinton, A. (2005). Breast-
feeding initiation in low-income women: role of attitudes, support, and perceived
control. Women's Health Issues, 15, 64-72.
Lanigan, J., Bishop, J., Kimber, A., & Morgan, J. P. (2001). Systematic review concerning the
age of introduction of complementary foods to the healthy full-term infant. European
Journal of Clinical Nutrition, 55, 309-320.
Louis, W., Davies, S., Smith, J., & Terry, D. (2007). Pizza and pop and the student
identity: the role of referent group norms in healthy and unhealthy eating.
Journal of Social Psychology, 147, 57-74.
McMillan, B., Conner, M., Woolridge, M., Dyson, G., Green, J., Renfrew M., et al. (2008).
Predicting breastfeeding in women living in areas of economic hardship: explanatory
role of the theory of planned behaviour. Psychology and Health, 23, 767-788.
National Health and Medical Research Council. (2003). Dietary guidelines for children
and adolescents incorporating the infant feeding guidelines for health workers.
Canberra: Commonwealth of Australia.
Naylor, A. J., & Morrow, A. L. (2001). Developmental readiness of normal full term
infants to progress from exclusive breastfeeding to the introduction of
complementary foods: Reviews of the relevant literature concerning infant
immunologic, gastrointestinal, oral motor and maternal reproductive and
lactational development. Washington D.C: Wellstart International and the
LINKAGES Project Academy for Educational Development.
Olson, B.H., Horodynski, M. A., Brophy-Herb, H., & Iwanski, K., C. (2008). Health
professionals’ perspectives on the infant feeding practices of low income mothers.
Maternal and Child Health Journal, 14, 75-85.
Rasmussen, K. M., & Kjolhede, C. L. (2007). Maternal obesity: a problem for both
mother and child. Obesity, 16, 929-931.
Rivis, A., & Sheeran, P. (2003). Descriptive norms as an additional predictor in the theory of
planned behaviour: a meta analysis. Current Psychology, 22, 218-233.
Schiess, S., Grote, V., Scaglioni, S., Luque, V., Martin, F., Stolarczyk, A., et al., (2010)
Introduction of complementary feeding in 5 European Countries. Journal Paediatric
Gastroenterology and Nutrition, 50, 92-98.
Scott, J. A., Binns, C. W., Graham, K. I., & Oddy, W. H. (2009). Predictors of the early
infant introduction of solids in infants: results of a cohort study [Electronic Version].
BMC Pediatrics, 9, 60 Retrieved from http://www.biomedcentral.com/1471-2431/9/60
Sheeran, P., Trafimow, D., & Armitage, C. J. (2003). Predicting behaviour from perceived
behavioural control: tests of the accuracy assumptions of the theory of planned
behaviour. The British Journal of Social Psychology, 42, 393-410.
Swanson, V., & Power, K. G. (2005). Initiation and continuation of breastfeeding:
theory of planned behaviour. Journal of Advanced Nursing, 50(3), 272-282.
Terry, D. J., & Hogg, M. A. (1996). Group norms and the attitude-behavior relationship: A
role for group identification. Personality and Social Psychology Bulletin, 22, 776-793.
Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987).
Rediscovering the social group: a self-categorisation theory. Oxford: Blackwell.
Verbeke, W., & Vackier, I. (2005). Individual determinants of fish consumption:
application of the theory of planned behaviour. Appetite, 44, 67-82.
White, K. M., Hogg, M. A., & Terry, D. J. (2002). Improving attitude-behavior
correspondence through exposure to normative support from a salient ingroup. Basic
and Applied Social Psychology, 24, 91-103.
White, K. M., Smith, J. R., Terry, D. J., Greenslade, J. H., & McKimmie, B. M. (2009).
Social influence in the theory of planned behaviour: the role of descriptive,
injunctive, and in-group norms. British Journal of Social Psychology, 48,
World Health Organisation, UNICEF. (2003). Global strategy for infant and young
child feeding. Geneva: WHO.
Wright, C. M., Parkinson, K. N., Drewett, R. F. (2004). Why are babies weaned early? Data
from a prospective population based cohort study. Archives of Disease in Childhood,
Means, standard deviations, and bivariate correlations for maternal and infant demographic variables, the TPB variables (attitude, subjective
norm, perceived behavioural control), intention (N=375) and reported behaviour (N=257) among first-time mothers.
1. Mother’s weight
-.05 .02 .08 .13* .09 .00 .01 -.02 -.04 -.24***
2. Mother’s education
- .32*** -.26*** .04 -.03 .03 -.01 .04 .07 .16*
3. Mother’s age
- -.10* -.01 -.01 .06 .07 .07 .13* .24***
4. Current feeding
- -.06 .03 -.12* -.03 -.10 -.16*** -.22***
5. Infant birth weight
- .07 -.04 .07 -.03 .03 -.01
(.92) .55*** .28*** .46*** .61*** .19**
7. Subjective norm
(.87**) .55*** .73*** .77*** .34***
(.64) .19*** .22*** .07
9. Group norm
- .63*** .28***
*p<.05, **p<.01, ***p<.001
Note. Mean scores on 7-point scales (1-7; higher scores stronger agreement, more important) except for mother’s self-reported pre-pregnancy
weight (1 = not overweight, 2 = overweight); mother’s education (1 = non-University, 2 = University); mother’s age at birth of infant (in years);
current infant feeding (1= breastfeeding, 2 = formula feeding at completion of first questionnaire, average infant age 13 weeks); infant birth
weight (in kilograms); and behaviour (age solid introduction, months). Note. The figures in brackets on the diagonal are alpha coefficients.
Where a construct was measured with two items, Pearson’s r (and significance) is reported. Note. PBC = perceived behavioural control.
Table 2. Hierarchical regression predicting first-time mothers’ intention to introduce solids at
6 months (N=358)
Infant birth weight
Infant birth weight
Infant birth weight
Note. R² = .038 for step 1; R² = .648 for step 2; R² = .653 for step 3.
* p < .05. *** p < .001.
Table 3. Hierarchical regression predicting first-time mothers’ actual behaviour of introducing
solids at 6 months (N=236)
Infant birth weight
Infant birth weight
Infant birth weight
Note. R² = .150 for step 1; R² = .319 for step 2; R² = .333 for step 3.
* p < .05. ** p < .01. *** p < .001.