Sage Journals: Discover world-class research

Abstract

Although research on the home literacy environment and its impact on early literacy has long focused on mothers, the past decade has seen a shift in scholarly attention to the role of fathers. Building on this shift, we examined whether the nature of parent–child interactions during shared storybook reading varies with parent gender, child gender and the interaction between the two, and we analysed whether possible differences in the nature of mother– and father–child interactions are related to story comprehension. We made video observations of mothers and fathers within 36 relatively highly educated families reading a storybook with their kindergartener (age 4 – 5) and registered the use of cognitively challenging (i.e. decontextualized) talk during these activities. After each shared reading session, we additionally administered a test assessing children’s understanding of the story being read. Two-way mixed ANOVA’s revealed no effects of parent gender or child gender on either the use of cognitively challenging talk or children’s story comprehension, nor did we find interaction effects of parent and child gender. The extent of cognitively challenging talk was significantly correlated to children’s comprehension scores for fathers, but not for mothers. This correlation seems to have masked another association, however: when correlations were computed separately for girls and boys, we found that the proportion of cognitively challenging utterances of both parents was correlated to comprehension scores for boys, but not for girls. The absence of parent gender effects provides further insights into the way mothers and fathers shape interactions during shared reading, but also stresses the need for studies with larger, more diverse samples. The observation that more frequent use of cognitively challenging talk was paralleled by better story comprehension for boys invites further research on the specific effects of shared reading for boys.

Keywords

Gender kindergarten children parent–child interactions picture books shared reading

Introduction

The extent to which children are engaged in literacy activities in their homes, often referred to as the home literacy environment (HLE), plays a crucial role in their early literacy development (Burgess et al., 2002; Leseman and de Jong, 1998; Niklas and Schneider, 2013; Van Steensel, 2006). Although research on the HLE and its impact on early literacy has long focused on mothers, the past decade has seen a shift in scholarly attention (Cutler and Palkovitz, 2020; Duursma, 2016). In line with changes in the caregiving roles of parents in western societies (Lamb, 2000), the number of studies on the contribution of fathers to the HLE has increased (Cutler and Palkovitz, 2020; Varghese and Wachen, 2015). In the current study, we build on this relatively new line of research by examining whether the nature of parent–child interactions during shared storybook reading varies with parent gender, child gender and the interaction between the two, and we analyse whether possible differences in the nature of mother– and father–child interactions are related to story comprehension.

Literature review

Research so far suggests that father–child literacy activities – particularly shared reading – have an important added value to children’s early literacy development. Duursma et al. (2008), for instance, found that Early Head Start fathers’ reported shared reading frequency at age three was significantly related to children’s receptive vocabulary at the same age, although the association was only found if fathers had at least a high school education. Fagan et al. (2016) also examined Head Start families and established that fathers’ shared reading frequency at age four was related to children’s concurrent language competence. While Duursma et al. (2008) and Fagan et al. (2016) focused on fathers’ contributions only, others compared the effects of mothers’ and fathers’ behaviours. Baker (2013), Baker (2014), for instance, found that both mothers’ and fathers’ home literacy involvement at age two (including shared reading frequency) uniquely predicted children’s literacy skills at age four, and that this was true in both Caucasian and African American families. Interestingly, research by Duursma (2014) suggests that fathers’ contributions might even be stronger than those by mothers: in a sample of Early Head Start families, she found that paternal shared reading frequency at age two predicted both concurrent and future vocabulary scores, whereas maternal shared reading frequency did not.

One possible explanation for the unique effects of father–child shared reading is that fathers engage in other types of interactions during shared reading than mothers do. More specifically, fathers have been suggested to use more ‘cognitively challenging’ talk: they are thought to employ more diverse vocabulary and ask more questions that require children to think beyond the literal text (Cutler and Palkovitz, 2020; Duursma, 2016). Two reasons have been proposed for this ‘differential experience’ (Bernstein-Ratner, 1988; Gleason, 1975; Korat et al., 2008; Rowe et al., 2004). On the one hand, it might be that because fathers generally engage in fewer shared activities with their children than mothers do, they might be less attuned to their children’s language level: while mothers adjust their interactions to their children’s emerging competence, fathers’ relative inexperience is hypothesized to make them use (too) complex talk. On the other hand, it might be that fathers approach shared reading differently from mothers: they might view shared reading as a teaching activity and consequently ‘impose larger demands on children and stretch their performance’ (Duursma, 2016: 479). In either case, fathers may be more inclined to talk to children in a grown-up fashion, forming a ‘bridge to the outside world’ (Gleason, 1975: 293).

A few studies have compared mother and father talk during shared reading, but results are mixed. Some researchers have indeed found that fathers use more complex language than mothers do. Korat et al. (2008) compared maternal and paternal mediation in storybook interactions with five- to six-year-olds in a diverse sample of low and middle socio-economic status (SES) families. They found that fathers used other, more challenging ways of expanding their children’s knowledge than mothers did: while most of the mothers’ extra-textual talk consisted of paraphrasing story fragments, fathers more often engaged in what Korat et al. refer to as ‘distancing’ (e.g. predicting what is going to happen next or drawing comparisons). Duursma (2016) compared mothers’ and fathers’ storybook readings with their preschoolers, focusing on the amount of ‘non-immediate talk’, which she defines as talk that goes beyond the information in the text or illustrations and, for instance, involves making predictions or connections to children’s experiences, drawing inferences or explaining events (Sigel, 1982). Both at age two and three, Duursma found that fathers used significantly more of such non-immediate talk than mothers. Finally, Vandermaas-Peeler et al. (2011) found an interesting interaction effect of parent and child gender on the complexity of interactions during shared reading. They compared the amount of ‘literacy-related guidance’ in mother and father talk, defined as a composite of the frequency of simple and elaborate descriptions, predictions, inferences and recalling text. Fathers were shown to provide twice as much of such guidance to girls than to boys, while for mothers there was no such difference.

Other researchers found no differences in cognitively challenging talk between mothers and fathers or they observed that mothers use more complex language than fathers do. Blake et al. (2006) compared mother and father talk during shared book reading in two age cohorts: one-year-olds and two-year-olds. They found that children’s age but not parent gender determined complexity level. For instance, both mothers and fathers asked more informative questions and questions about children’s feelings to older children than to younger children and both parent genders did this to a similar extent. Schwartz (2004) examined behaviours of mothers and fathers during shared reading with their two-year-olds. She found that fathers used significantly more literal and, thus, less challenging questions than mothers (e.g. questions asking children to label or localize an object on a picture), while mothers used more ‘expansions’, that is, instances where they elaborated on children’s utterances, for example, by making connections with children’s previous experiences or world knowledge.

Although conclusions on the similarities and differences between mother– and father–child interactions during shared reading remain tentative, the quality of father talk does appear to make a unique contribution to children’s language outcomes. Blake et al. (2006), for instance, found positive and significant correlations between fathers’ use of complex questions with their one- and two-year-olds and children’s scores on the McArthur Communicative Development Inventory. Pancsofar et al. (2010) found that fathers’ use of diverse vocabulary during shared reading to their infants predicted children’s communicative skills at 15 months and their expressive language scores 2 years later. Malin et al. (2014) found that the amount of ‘metalingual talk’ fathers produced during shared reading interactions with their two-year-olds predicted children’s pre-kindergarten receptive vocabulary scores. Finally, Baker et al. (2015) registered fathers’ mean length of utterance (MLU) during shared reading as a measure of linguistic complexity and found that MLU assessed before kindergarten entry positively predicted children’s receptive vocabulary scores during kindergarten. Interestingly, in both the studies by Pancsofar et al. (2010) and Baker et al. (2015), the quality of mothers’ talk did not predict language outcomes when entered in a model with fathers’ quality of talk. The latter might imply that fathers better succeed in making shared reading a learning experience than mothers do: their possibly more frequent use of cognitively challenging talk could have positive consequences for what children take in from a story (i.e. it may help them learn more words or better process the storyline). To test whether this is a viable hypothesis, additional research linking mother– and father–child shared reading interactions to comprehension of the story being read is required.

As we have argued, examining the role of fathers in children’s literacy development is a relatively new line of inquiry. Still, it is based on a rather traditional perspective on both gender and family composition: it assumes that families include a mother and a father, whose identities are clearly distinguishable. However, gender identities are not necessarily clear cut (Bem, 1974; Spence, 1993) and there are various non-traditional family configurations (e.g. single-parent families, same-sex parent families and extended families). From our review of the literature, we must conclude that such issues are hardly touched upon in home literacy research (Cutler and Palkovitz, 2020). While gender identity has been found to be a predictor of reading motivation in some recent studies (Espinoza and Strasser, 2020; McGeown et al., 2015; McGeown, 2012), how gender identity steers the role different caretakers play in children’s literacy development is as yet unclear. Similarly, little systematic research has been done on home literacy patterns in non-traditional families (Prins et al., 2021). Our aim in this study is to add to the small, but growing research base comparing mothers’ and fathers’ contributions to early literacy. Similar to the above-described studies comparing mothers and fathers, we designed our study from a more traditional perspective on gender and family. Nevertheless, we acknowledge that family contexts have become increasingly diverse.

The current study

The frequency and quality of father–child shared reading activities appear to make a unique contribution to children’s early language outcomes. The question whether this is caused by special features of father–child interactions during shared reading remains unanswered: while some studies comparing mothers’ and fathers’ storybook reading show differences in the complexity of parent–child talk, others do not. In their recent review, Cutler and Palkovitz (2020) therefore call on researchers to extend the research base that examines parental distinctions in shared reading styles. In the current study, we follow this call and compare mothers’ and fathers’ use of cognitively challenging talk during shared reading. Following Vandermaas-Peeler et al. (2011) observations, we additionally examine whether there is an interaction effect of parent and child gender, that is, whether parents engage in different types of interactions with girls than with boys and whether possible differences are dependent on parent gender. We also assess whether differences in the use of cognitively challenging talk between mothers and fathers have implications for what children learn from stories. Assuming there are such differences, there are two possible outcomes. On the one hand, it may be that fathers’ more frequent use of cognitively challenging talk supports children’s story comprehension because it elicits deeper processing. On the other hand, it may be that fathers’ talk is too challenging and thus interferes with children’s understanding of the story being read. Consequently, we test whether children’s story comprehension differs when being read to by mothers and fathers. We also assess whether the use of cognitively challenging talk is related to story comprehension, and whether these relations differ dependent on parent and child gender.

These aims result in four research questions:

(1) Do mothers and fathers differ in the use of cognitively challenging talk during shared reading interactions with their children?

(2) Do differences between mothers’ and fathers’ use of cognitively challenging talk depend on the gender of the child?

(3) Do children comprehend stories better after being read to by fathers than by mothers? And is there a difference between boys and girls?

(4) Are there differences between mothers and fathers in how the use of decontextualized language plays a role in story comprehension? And what role does child gender play?

Method

Design

We made video observations of mothers and fathers within families reading a storybook with their kindergartener (age 4 – 5) and registered the use of cognitively challenging talk during these activities. After each shared reading session, we additionally administered a test assessing children’s understanding of the story being read.

Participants

Thirty-six families participated in the study, each with one mother or female caregiver and one father or male caregiver. In one family, the caregivers were the child’s grandparents, who had taken over the care of the natural parents (sensitivity analyses without this family showed that for all research questions outcomes were similar to those including this family). Families were recruited in the personal and professional networks of the second, third and fourth author. A first inclusion criterion was that families had to include a mother/female caregiver and a father/male caregiver: to be able to infer parent gender effects, child effects had to be kept constant and thus mothers and fathers had to come from the same families. Note that this did not necessarily imply the exclusion of same-sex families: for instance, a family with two mothers and one father might have been included. A second inclusion criterion was that both parents or caregivers (hence: parents) regularly (i.e. at least once a month) engaged in shared reading activities with the child. Table 1 presents the characteristics of participating children and parents.

Table 1.

Participant characteristics.

Children (n = 36)
Gender	Girl	21 (58%)
Gender	Boy	14 (42%)
Age	Overall	M = 5; 3 (SD = 0; 6)
	Girls	M = 5; 3 (SD = 0; 6)
	Boys	M = 5; 3 (SD = 0; 6)
Parents (n = 72)
Gender	Female	36 (50%)
Gender	Male	36 (50%)
Age	Overall	M = 39; 2 (SD = 6; 2)
	Mothers	M = 38; 0 (SD = 4; 9)
	Fathers	M = 40; 5 (SD = 7; 2)
Country of birth	Netherlands	64 (89%)
Country of birth	Other^a	8 (11%)
Educational level	Secondary education	5 (7%)
	Intermediate vocational training	9 (13%)
	Higher vocational training	33 (46%)
	University	25 (35%)
Shared reading frequency	Every day	40 (56%)
	Once/few times a week	29 (40%)
	Once/few times a month	3 (4%)
Families (n = 36)
Home language	Dutch	31 (86%)
Home language	Dutch and another language^b	5 (14%)

Note: ^a Other birth countries of parents were Belgium (n = 3), Germany (n = 1), Indonesia (n = 1), Netherlands Antilles (n = 1), Peru (n = 1) and Surinam (n = 1).

^b Other home languages were Arabic (n = 1), English (n = 1), German (n = 1), Spanish (n = 1) and Turkish (n = 1).

Measures

Observations

During the observations, parents were asked to read one of two picture books with their child. Each picture book contained twelve pages, with about four to five lines of text and one corresponding drawing per page. The first book, Eekhoorn en Ekster gaan op zoek naar eikels (‘Squirrel and Magpie go looking for acorns’) was 388 words long and described how Squirrel invites his friend Magpie to search for acorns. Because Magpie is more interested in shiny things, he overlooks all kinds of special situations (e.g. a hedgehog with socks on his spines). At the end of the story, Magpie realizes that due to his inattention he missed a lot of beautiful things. The second book, Het geluid van Eend (‘The sound of Duck’) was 450 words long and described the story of Duck, who wants to make friends with the animals on a farm by mimicking their sounds. However, the animals laugh at her because her imitations fail. Then she meets Frog, who encourages Duck to make her own sound (i.e. quacking). When the other animals hear Duck and Frog ‘singing’ together (in Dutch, representations of sounds that ducks and frogs make are similar, i.e. kwaak), they come over to listen and Duck no longer feels sad and lonely. Both books had been developed for a previous study (De Koning et al., 2020), which ensured that families had not seen the books before.

Observations were video-taped, and all extra-textual utterances were transcribed and coded. We used an existing coding scheme (De Koning et al., 2020; De la Rie et al., 2018; De la Rie et al., 2020), that allows to determine how cognitively challenging parent–child interactions are by coding the level of ‘decontextualization’ for every utterance (Curenton et al., 2008; Snow, 1983). The scheme was adapted from schemes used by Van Kleeck et al. (1997) and Sorsby and Martlew (1991), which were, in turn, based on the distinction in decontextualization levels proposed by Blank et al. (1978).

Utterances in the transcripts were distinguished by ‘unit of meaning’: a thematically consistent verbalization of a single speech act (De Backer et al., 2012; De Koning et al., 2020; De la Rie et al., 2018; De la Rie et al., 2020). When an utterance or a series of utterances could be coded with one single code, it was seen as one unit of meaning. Utterances were first coded as ‘content-related’ (i.e. utterances related to the content of the story) or ‘other’ (i.e. feedback and acknowledgement, procedural, off-task, ambiguous or inaudible). Content-related utterances were further coded into four levels of decontextualization (see Table 2 for sample utterances). Utterances were coded at Level 1 (‘matching perception’) if they were limited to the directly perceivable context of the story. These utterances involve labelling (1a), locating (1b) and noticing people, animals or objects in the storybook (1c). Utterances at Level 2 (‘selective analysis and integration of perception’) were limited to the directly perceivable context as well but placed a slightly higher representational demand on children. These utterances involve questions or comments that require the child to focus on specific aspects of objects and characters (2a), or on events and actions (2b), or to finish a sentence (2c). Utterances at Level 3 (‘reorder/infer about perception’) move beyond the immediate context of the story. They involve making connections between different parts of the story (3a), defining words (3b), making comparisons to the child’s own experiences (3c) and discussing personal opinions (3d). Level 4 utterances (‘reasoning about perception’) are defined as the highest level of decontextualization. They involve making predictions (4a) and drawing conclusions (4b). Similar to previous studies (De Koning et al., 2020; De la Rie et al., 2018; De la Rie et al., 2020), we defined cognitively challenging talk as utterances at Levels 3 and 4.

Table 2.

Translated sample utterances per category (parent utterances in bold).

Category	Example
Level 1: ‘Matching perception’
1a: Labelling	Mother: ‘Who is this?’
1a: Labelling	Child: ‘Frog’.
1b: Locating	Mother: ‘And where do you see any more socks?’ (Child points at a sock near hedgehog)
1c: Noticing	Mother (referring to an elephant that is standing on its front legs): ‘Did you already see it?’
	Child: ‘Yes’.
	Mother: ‘Yes?’ (Child points at the picture of the elephant)
Level 2: ‘Selective analysis and integration of perception’
2a: Focusing on aspects of objects/characters	Mother: ‘Is that a big basket or a small basket?’
2a: Focusing on aspects of objects/characters	Child: ‘Small basket’.
2b: Focusing on events and actions	Mother: ‘What is Magpie looking for? What is he looking for?’
2b: Focusing on events and actions	Child: ‘Shiny things’.
2c: Finishing a sentence	Father: ‘The sound of a …’
2c: Finishing a sentence	Child: ‘… Duck’.
Level 3: ‘Reorder/infer about perception’
3a: Making connections between different parts of the story	Mother: ‘You indeed said that she could not quack, what is she actually not able to do?’ (Mother shows the book cover and makes the connection with the scene they just read)
3a: Making connections between different parts of the story	Child: ‘Quoo, quoo, quoo’ (child makes a sound that combines quacking and mooing)
3b: Defining words	Father: ‘Do you know what lonely is?’
3b: Defining words	Child: ‘Yes, that you … yes that it is not enjoyable, she wants to play with someone’.
3c: Making comparisons to child’s own experiences	Father: ‘Have you ever seen those? Those shiny things?’
3c: Making comparisons to child’s own experiences	Child: ‘Yes, with grandma Mieke’.
3d: Discussing personal opinions	Father: ‘That is kind of special, isn’t it, an elephant making a handstand just like that?’
3d: Discussing personal opinions	Child: ‘Yes and that’s very skillful as well’.
Level 4: ‘Reasoning about perception’
4a: Making predictions	Mother: ‘What do you think the story will be about?’
4a: Making predictions	Child: ‘Uhm, that they are going to look where the sound of Duck is’.
4b: Drawing conclusions	Mother reads aloud: But Magpie does not pay any attention to Squirrel
	Mother: ‘Why not, do you think? What could he be doing?’
	Child: ‘Only looking for shiny things’.

Story comprehension

For each of the two picture books, we constructed a comprehension test to assess children’s understanding of the story content. We based the tests on Paris and Paris’ (2003) Narrative Comprehension of Picture Books task. We asked children ten open-ended comprehension questions about each book. Five were about explicit information in the text and addressed the story characters, setting, initiating event, problem and outcome resolution. The other five required children to make inferences based on the characters’ feelings, dialogues, causal relations, predictions and themes. To exemplify what questions were asked, the Appendix includes one of the tests. Questions were read aloud by a researcher. In case a question referred to a picture in the picture book, the researcher showed the corresponding picture. Children’s answers to the questions were audio-recorded and transcribed. From the transcriptions, answers were scored according to rubrics that contained all correct answers and their corresponding scores. Per question, children were awarded 0, 1 or 2 points. A score of 0 points indicated a wrong answer or no answer; one point represented a correct answer; two points were awarded if the correct answer was accompanied by a coherent explanation. For each picture book separately, an overall comprehension score was computed by summing the scores on all 10 questions (ranging from 0 to 20 points) about the picture book.

Procedure

Observations and test administrations were conducted in families’ homes by Authors 2, 3 and 4. During visits of approximately 1 hour, a researcher asked both mother and father to read a designated storybook (either ‘Squirrel and Magpie go looking for acorns’ or ‘The sound of Duck’). Which picture book each parent read and the order in which they did this were counterbalanced across parent–child pairs to eliminate confounding effects of picture book and order. Each parent–child pair was therefore randomly assigned to one of four situations (see Table 3).

Table 3.

Counterbalancing parent–child pairs across different situations.


Situation 1	Situation 2
Session 1: Father reads Squirrel and Magpie	Session 1: Mother reads Squirrel and Magpie
Session 2: Mother reads duck	Session 2: Father reads duck
Situation 3	Situation 4
Session 1: Father reads Duck	Session 1: Mother reads Duck
Session 2: Mother reads Squirrel and Magpie	Session 2: Father reads Squirrel and Magpie

The researcher followed a standardized protocol. First, to make the child feel at ease, the researcher introduced herself and started a brief conversation, for instance, about what the child was wearing or what the child did at school that day. Subsequently, she briefly explained the research and the procedure, after which the parents were asked to sign a consent form. Then, the researcher prepared the observation session: she put the picture books in place and set up the video camera and/or voice-recorder. Since observations are not a natural setting for children, children were actively involved at this stage: children were invited to make a test recording. When the preparations were finished, the researcher asked the first parent to start the first reading session, instructing her or him to read the book as she or he would normally do, and started the recording. During the session, the researcher stayed in the room but sat apart from the parent and child, maintaining a neutral stance as much as possible (i.e. by not responding to child and parent during the activity, neither verbally nor non-verbally). The other parent was urged to distance her- or himself, to prevent (s)he would observe the other parent and (unconsciously) take over her or his behaviour. After parent and child indicated the session was finished, the researcher administered the story comprehension task. The parent was asked to leave to ensure that she or he could not (un)consciously influence the child’s answers. However, if this made the child feel uncomfortable, the parent was allowed to stay and was asked not to interfere with the test administration. The researcher was instructed to give positive feedback to the child without helping the child to answer the questions (e.g. ‘I can see you are doing your best!’). After the first observation and test administration, the procedure was repeated with the second parent. At the conclusion of the visit, the child was given a small present (a sticker book or colouring book) and the parents were handed a gift certificate of 12.5 euros. Video and audio recordings were stored on SURFdrive, a safe and university-approved cloud system.

The transcripts of the observations were coded by three trained coders (Authors 2, 3 and 4). Coders were trained during two sessions. In Session 1, the first author explained the coding scheme using a sample transcript. Between Sessions 1 and 2, the coders independently coded a practice transcript. The first author then computed the agreement between each coder and the model coding, using Krippendorff’s α (Hayes and Krippendorff, 2007). Both for content-related versus other utterances (α = 0.69) and for level of decontextualization (α = 0.72), interrater reliabilities were above the norm of α = 0.67. All 72 transcripts were then double coded. We applied ‘blind’ coding to limit bias: for instance, when Author 2 had conducted the observation in a family, Authors 3 and 4 coded the transcripts of this family. The average interrater reliabilities were above the norm (content-related versus other utterances: α = 0.75; level of decontextualization: α = 0.76). Finally, each pair of coders (Authors 2–3, Authors 3–4 and Authors 2–4) discussed and resolved discrepancies to determine the final codings.

The story comprehension tests were scored by Authors 2, 3 and 4 as well. Each test was assessed by a pair of raters (Authors 2–3, Authors 3–4 and Authors 2–4), once again using an unbiased, blind procedure so that none of the researchers assessed a test she had administered. The average interrater reliability between coders was high (Krippendorff’s α = 0.82). Discrepancies between each pair’s scores were discussed and resolved to determine the final scores.

Analysis

To answer Research Questions 1 and 2, we used a two-way mixed ANOVA with parent gender as the within-subjects factor and child gender as the between-subjects factor. Similar to previous studies (De Koning et al., 2020; De la Rie et al., 2018; De la Rie et al., 2020), we defined the dependent variable – cognitively challenging talk – as the proportion of the total number of utterances that were at Level 3 or 4. This analysis allowed us to determine whether there were differences between mothers and fathers in their use of cognitively challenging talk, whether parents varied in their use of cognitively challenging talk in interactions with girls and boys, and whether possible differences in the use of cognitively challenging talk in interactions with girls and boys were dependent on parent gender. Inspection of the boxplot indicated two outliers but because there were no substantive reasons for excluding these cases and because the studentized residuals had no values greater than ± 3, we decided to keep the sample intact. Neither the Shapiro–Wilk test (ps > 0.05) nor the Normal Q-Q plot showed deviations from nonnormality. There was homogeneity of variances, as assessed by Levene’s test (ps > 0.05). Finally, Box’s test indicated there was homogeneity of covariances (p = .964).

To answer Research Question 3, we also conducted a two-way mixed ANOVA with parent gender as the within-subjects factor and child gender as the between-subjects factor, but with the story comprehension task as dependent variable. Neither the boxplot nor the studentized residuals indicated the presence of outliers. Neither the Shapiro–Wilk test (ps > 0.05) nor the Normal Q-Q plot showed deviations from nonnormality. There was both homogeneity of variances (Levene’s test: all ps > 0.05) and homogeneity of covariances (Box’s test: p = .672).

To answer Research Question 4, we first computed overall correlations between the proportion of Level 3 and 4 utterances of mothers and fathers and children’s scores on the comprehension test. Subsequently, we computed these correlations separately for girls and boys.

Results

Descriptive statistics

Table 4 presents descriptive statistics for the total numbers of utterances by mothers and fathers, the numbers of Level 3 and 4 utterances and the proportions of Level 3 and 4 utterances, both overall and broken down into utterances directed to girls and boys. The table also presents the descriptive statistics for the comprehension test.

Table 4.

Descriptive statistics.

	Range	Mean (SD.)
Total number of utterances mother	2–65	26.19 (16.23)
To girls	2–65	29.76 (17.50)
To boys	4–50	21.20 (13.23)
Total number of utterances father	3–49	20.14 (13.12)
To girls	3–49	20.43 (13.28)
To boys	3–44	19.73 (13.34)
Number of level 3 and 4 utterances mother	0–22	6.50 (5.59)
To girls	0–22	7.76 (5.80)
To boys	0–15	4.73 (4.94)
Number of level 3 and 4 utterances father	0–16	4.17 (3.88)
To girls	0–16	4.38 (4.11)
To boys	0–13	3.87 (3.66)
Proportion of level 3 and 4 utterances mother	0.00–.59	0.23 (0.13)
To girls	0.00–.59	0.24 (0.12)
To boys	0.00–.50	0.21 (0.14)
Proportion of level 3 and 4 utterances father	0.00–.48	0.19 (0.12)
To girls	0.00–.48	0.21 (0.12)
To boys	0.00–.38	0.16 (0.12)
Comprehension test	1.00–15.50	9.17 (3.30)

Mothers used more utterances than fathers did and, as a likely consequence, also more Level 3 and 4 utterances. Conclusions on whether mothers and fathers differ in their use of cognitively challenging talk might thus be distorted by the fact that mothers overall talked more during shared reading than fathers did. Consequently, we decided to base further analyses on proportions of Level 3 and 4 utterances (for a similar procedure, see De la Rie et al., 2018; De la Rie et al., 2020). The mean score on the comprehension test indicates that, on average, children had about half of the questions correct.

Research Questions 1 and 2

The two-way mixed ANOVA showed that there were no main effects of either parent gender (F [1, 34] = 2.92, p = .097) or child gender (F [1, 34] = 1.06, p = .310) on the proportion of Level 3 and 4 utterances produced by parents. There was also no significant interaction effect of parent and child gender (F [1, 34] = 0.00, p = .964). In other words, mothers and fathers did not differ in the extent to which they used cognitively challenging talk in shared reading interactions with their child; whether the child was a girl or boy also did not make a difference.

Research Question 3

The two-way mixed ANOVA showed that there were no main effects of either parent gender (F [1, 34] = 0.57, p = .457) or child gender (F [1, 34] = 2.21, p = .147) on the scores on the comprehension test. There was also no significant interaction effect of parent and child gender (F [1, 34] = 0.15, p = .698). In other words, variability in children’s comprehension scores was not dependent on whether they had been read to by their mother or father; whether a child was a girl or boy did not matter either.

Research question 4

The overall correlation between the proportion of Level 3 and 4 utterances by parents and children’s comprehension scores was r = 0.30, p = .001: the larger the share of cognitively challenging utterances, the better children’s understanding of the stories. For mothers, the correlation between the proportion of Level 3 and 4 utterances and children’s comprehension scores was not significant (r = 0.28, p = .099), but for fathers, it was (r = 0.38, p = .023). Subsequently, these correlations were computed separately for girls and boys. For girls, neither correlation with comprehension scores was statistically significant (mothers’ proportion of Level 3 and 4 utterances: r = 0.07, p = .774; fathers’ proportion of Level 3 and 4 utterances: r = 0.18, p = .435). For boys, however, both correlations were significant (mothers’ proportion of Level 3 and 4 utterances: r = 0.56, p = .029; fathers’ proportion of Level 3 and 4 utterances: r = 0.56, p = .030).

Discussion

Although mothers tended to talk more during shared reading activities than fathers, mothers and fathers did not differ in the extent to which they used cognitively challenging talk: the proportions of Level 3 and 4 utterances in parent–child interactions were similar for both genders (Research Question 1). Child gender did not play a role either: parents did not use more cognitively challenging talk in interactions with girls than with boys and there were no interaction effects of parent and child gender (Research Question 2). Additionally, there were no parent gender, child gender or interaction effects on story comprehension (Research Question 3). The share of cognitively challenging talk was associated with story comprehension: more Level 3 and 4 utterances by parents were paralleled by a better understanding by children of the stories being read. This association appeared to be related to parent gender: for fathers the correlation between proportion of Level 3 and 4 utterances and children’s comprehension scores was significant, for mothers it was not. This correlation seems to have masked another association, however: when correlations were computed separately for girls and boys, we found that the proportion of cognitively challenging utterances of both parents was correlated with comprehension scores for boys, but not for girls (Research Question 4).

The absence of effects of parent gender on the share of cognitively challenging utterances may be explained by the characteristics of our sample: most parents were higher educated and read to their children at least once a week and this was true for both mothers and fathers. Whereas the notion of differential experiences in shared reading is partly based on fathers’ assumed lack of experience (Korat et al., 2008; Rowe et al., 2004), this probably not holds for the fathers in our sample: most of them were likely practiced readers of children’s books. There are other possible explanations as well. We examined mother–father pairs: each pair thus read to the same child. Since characteristics of the child (e.g. her/his language ability) influence the nature of parent–child interactions (Saracho and Spodek, 2010), similarities in interaction patterns between mothers and fathers may reflect a child factor: specific characteristics of the child determine how both parents respond. Parents within a family may also learn from one another. The Family Systems Theory, for instance, suggests that mothers and fathers tend to adjust their parenting styles (Cox and Paley, 1997). This may also hold for shared reading: mothers and fathers may observe each other while reading and take over one another’s behaviours.

The observation that mothers’ and fathers’ use of Level 3 and 4 utterances was related to children’s comprehension scores may be explained in two ways. On the one hand, it may imply that parents better succeed in discussing picture books more elaborately when children have a better understanding of the story being read: better story comprehension might allow more cognitively challenging interactions. On the other hand, it could mean that, when parents engage their child in more challenging interactions, they help her or him understand the story better. Following the latter interpretation, it was interesting to see that correlations between the proportion of Level 3 and 4 utterances and story comprehension were significant for boys, but not for girls. Possibly, boys needed more support in understanding the stories than girls or, rather, they needed to be challenged more. This interpretation aligns with the outcomes of previous research on young children’s literacy interests. In a study among five-year-old children, Meagher et al. (2008), for instance, found that boys showed significantly less engagement during shared reading than girls did. Baroody and Diamond (2013) found that both parents and teachers of four- and five-year-olds reported significantly less interest in literacy activities such as shared reading for boys than for girls.

Limitations and future directions

One limitation of our study is that there was relatively little diversity in our sample: most parents were high educated and were born in the Netherlands. Interestingly, previous studies that included low SES families did find differences between mothers and fathers in interaction patterns (Duursma, 2016; Korat et al., 2008). Therefore, a question that deserves further scrutiny is to which extent SES and/or other variables, such as ethnic background, are responsible for the presence or absence of gender effects. From the afore-mentioned studies, only the study by Korat et al. allowed a comparison of SES groups – their sample included 19 low SES and 20 middle SES families – but they only tested main effects of parent gender and SES, not their interaction, possibly due to the relatively small sample size. To be able to analyse interaction effects of parent gender and characteristics such as SES and ethnic background, we therefore call for research based on larger, more diverse samples.

The sample is homogeneous in another way as well: apart from one family in which the child was raised by grandparents, the sample did not include any non-traditional families (e.g. single-parent families and families with same-sex parents). Although our design did not necessarily rule out the participation of these families (see Participants), we did not come across these families during recruitment. Inclusion of non-traditional families would enrich our knowledge of how home literacy environments take shape. Extending research designs to include information on gender identities might prove insightful as well, as previous studies on reading motivation have shown (Espinoza and Strasser, 2020; McGeown et al., 2015; McGeown, 2012).

Another limitation is that we cannot draw definitive conclusions about the causal mechanism underlying the association between the use of cognitively challenging talk and story comprehension for boys. It could be that the use of more challenging, decontextualized talk helps in engaging boys during shared reading and that this has positive consequences for their processing of text. To test the validity of this assumption, intervention studies would be necessary. Interestingly, however, we could not find any studies that analysed effects of interventions such as Dialogic Reading – in which cognitively challenging parent–child interactions are promoted – separately for girls and boys (e.g. Mol et al., 2008). We therefore recommend conducting experimental studies that assess whether introducing cognitively challenging talk during shared reading activities with boys results in higher attention, stronger engagement and more favourable language outcomes.

A final limitation is that we focused on parental utterances only. The positive correlation between the proportion of Level 3 and 4 utterances and comprehension scores suggests parental utterances help children in processing stories more actively (see previous Limitation). However, sequential analysis of utterances by parents and children and the connections between those utterances might have shed even more light on the cognitively challenging nature of the interactions.

Conclusion

Our study provides further insight into the effects of parent and child gender on the nature of parent–child interactions during shared reading. The fact that we found no effects of parent gender on either the use of cognitively challenging talk or children’s story comprehension suggests that, at least in higher SES families where both parents are engaged readers, mothers and fathers are more alike than is assumed (Cutler and Palkovitz, 2020; Duursma, 2016). The contrast with the outcomes of previous studies using other types of samples (Duursma, 2016; Korat et al., 2008) stresses the need for studies with larger, more diverse samples, so that interaction effects of parent and child gender and background characteristics can be tested. The observation that more frequent use of cognitively challenging talk was paralleled by better story comprehension for boys invites further research on the specific effects of shared reading for boys.

Footnotes

Acknowledgements

We would like to thank the families taking part in this study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Notes

Brenda Gouw is alumnus of Erasmus University Rotterdam, Saskia Liefers and Tessa van Aspert are alumni of Free University Amsterdam.

ORCID iD

Roel van Steensel

Appendix

References

Baker

(2013) Fathers’ and mothers’ home literacy involvement and children’s cognitive and social emotional development: implications for family literacy programs. Applied Developmental Science 17(4): 184–197.

Baker

(2014) African American fathers’ contributions to children’s early academic achievement: evidence from two-parent families from the early childhood longitudinal study-birth cohort. Early Education and Development 25(1): 19–35.

Baker

Vernon-Feagans

Life, The Family Project Investigators (2015) Fathers’ language input during shared book activities: Links to children’s kindergarten achievement. Journal of Applied Developmental Psychology 36: 53–59.

Baroody

Diamond

(2013) Measures of preschool children’s interest and engagement in literacy activities: measures of preschool children's interest and engagement in literacy activities: examining gender differences and construct dimensionsearly childhood research quarterly. Early Childhood Research Quarterly 28(2): 291–301.

Bem

(1974) The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology 42(2): 155–162.

Ratner

(1988) Patterns of parental vocabulary selection in speech to very young children. Journal of Child Language 15(3): 481–492.

Blake

Macdonald

Bayrami

, et al. (2006) Book reading styles in dual-parent and single-mother families. The British Journal of Educational Psychology 76(3): 501–515.

Blank

Rose

Berlin

(1978) The Language of Learning: The Preschool Years. New York, NY: Grune & Stratton.

Burgess

Hecht

Lonigan

(2002) Relations of the home literacy environment (HLE) to the development of reading-related abilities: a one-year longitudinal study. Reading Research Quarterly 37(4): 408–426.

10.

Cox

Paley

(1997) Families as systems. Annual Review of Psychology 48(1): 243–267.

11.

Curenton

Craig

Flanigan

(2008) Use of decontextualized talk across story contexts: how oral storytelling and emergent reading can scaffold children’s development. Early Education and Development 19(1): 161–187.

12.

Cutler

Palkovitz

(2020) Fathers’ shared book reading experiences: common behaviors, frequency, predictive factors, and developmental outcomes. Marriage & Family Review 56(2): 144–173.

13.

De Backer

Van Keer

Valcke

(2012) Exploring the potential impact of reciprocal peer tutoring on higher education students’ metacognitive knowledge and regulation. Instructional Science 40(3): 559–588.

14.

De Koning

Wassenburg

Ganushchak

, et al. (2020) Inferencing questions embedded in a children’s book help children make more inferences. First Language 40(2): 172–191.

15.

De la Rie

Van Steensel

Van Gelderen

AJS

, et al. (2018) The role of type of activity in parent-child interactions within a family literacy program: comparing prompting boards and shared reading. Early Child Development and Care 188(8): 1076–1092.

16.

De la Rie

Van Steensel

Van Gelderen

AJS

, et al. (2020) Level of abstraction in parent–child interactions: the role of activity type and socioeconomic status. Journal of Research in Reading 43(1): 140–159.

17.

Duursma

(2014) The effects of fathers’ and mothers’ reading to their children on language outcomes of children participating in early head start in the United States. Fathering 12(3): 283–302.

18.

Duursma

(2016) Who does the reading, who the talking? Low-income fathers and mothers in the US interacting with their young children around a picture book. First Language 36(5): 465–484.

19.

Duursma

Pan

Raikes

(2008) Predictors and outcomes of low-income fathers’ reading with their toddlers. Early Childhood Research Quarterly 23(3): 351–365.

20.

Espinoza

Strasser

(2020) Is reading a feminine domain? The role of gender identity and stereotypes in reading motivation in Chile. Social Psychology of Education 23: 861–890.

21.

Fagan

Iglesias

Kaufman

(2016) Associations among head start fathers’ involvement with their preschoolers and child language skills. Early Child Development and Care 186(8): 1342–1356.

22.

Gleason

(1975) Fathers and other strangers: men’s speech to young children. In: Dato

IDP

(ed), Developmental Psycholinguistics: Theory and Applications. Washington, DC: Georgetown University Press, pp. 289–297.

23.

Hayes

Krippendorff

(2007) Answering the call for a standard reliability measure for coding data. Communication Methods and Measures 1(1): 77–89.

24.

Korat

Ron

Klein

(2008) Cognitive mediation and emotional support of fathers and mothers to their children during shared book-reading in two different SES groups. Journal of Cognitive Education and Psychology 7(2): 223–247.

25.

Lamb

(2000) The history of research on father involvement. Marriage and Family Review 29(2–3): 23–42.

26.

Leseman

PPM

Jong

(1998) Home literacy: opportunity, instruction, cooperation and social-emotional quality predicting early reading achievement. Reading Research Quarterly 33(3): 294–318.

27.

Malin

Cabrera

Rowe

(2014) Low-income minority mothers’ and fathers’ reading and children’s interest: longitudinal contributions to children’s receptive vocabulary skills. Early Childhood Research Quarterly 29: 425–432.

28.

McGeown

(2012) Sex or gender identity? Understanding children’s reading choices and motivation. Journal of Research in Reading 38(1): 35–46.

29.

McGeown

Goodwin

Henderson

, et al. (2015) Gender differences in reading motivation: does sex or gender identity provide a better account?. Journal of Research in Reading 35(3): 328–336.

30.

Meagher

Arnold

Doctoroff

, et al. (2008) The relationship between maternal beliefs and behavior during shared reading. Early Education and Development 19(1): 138–160.

31.

Mol

Bus

De Jong

, et al. (2008) Added value of dialogic parent-child book readings: a meta-analysis. Early Education and Development 19: 7–26.

32.

Niklas

Schneider

(2013) Home literacy environment and the beginning of reading and spelling. Contemporary Educational Psychology 38(1): 40–50.

33.

Pancsofar

Vernon-Feagans

Life, The Family Project Investigators (2010) Fathers’ early contributions to childrens language development in families from low-income rural communities. Early Childhood Research Quarterly 25: 450–463.

34.

Paris

(2003) Assessing narrative comprehension in young children. Reading Research Quarterly 38(1): 3676.

35.

Prins

Clymer

Kaiper-Marquez

, et al. (2021) Family literacy. In: Rocco

Smith

Mizzi

, et al. (eds) The Handbook of Adult and Continuing Education. Sterling, VA: Stylus.

36.

Rowe

Coker

Pan

(2004) A comparison of fathers’ and mothers’ talk to toddlers in low-income families. Social Development 13(2): 278–291.

37.

Saracho

Spodek

(2010) Parents and children engaging in storybook reading. Early Child Development and Care 180(10): 1379–1389.

38.

Schwartz

(2004) An observational study of mother/child and father/child interactions in story reading. Journal of Research in Childhood Education 19(2): 105–114.

39.

Sigel

(1982) The relationship between parental distancing strategies and the child’s cognitive behavior. In: Laosa

Sigel

(eds) Families as Learning Environments for Children. Boston, MA: Springer, pp. 47–86.

40.

Snow

(1983) Literacy and language: relationships during the preschool years. Harvard Educational Review 53(2): 165–189.

41.

Sorsby

Martlew

(1991) Representational demands in mothers’ talk to preschool children in two contexts: picture book reading and a modelling task. Journal of Child Language 18(2): 373–395.

42.

Spence

(1993) Gender-related traits and gender ideology: evidence for a multifactorial theory. Journal of Personality and Social Psychology 64: 624–635.

43.

Vandermaas-Peeler

Sassine

Price

, et al. (2011) Mothers’ and fathers’ guidance behaviours during storybook reading. Journal of Early Childhood Literacy 12(4): 415–442.

44.

Van Kleeck

Gillam

Hamilton

, et al. (1997) The relationship between middle-class parents’ book-sharing discussion and their preschoolers’ abstract language development. Journal of Speech, Language, and Hearing Research 40(6): 1261–1271.

45.

Van Steensel

(2006) Relations between socio-cultural factors, the home literacy environment and children’s literacy development in the first years of primary education. Journal of Research in Reading 29(4): 367–382.

46.

Varghese

Wachen

(2015) The Determinants of father involvement and connections to children’s literacy and language outcomes: review of the literature. Marriage & Family Review 52(4): 331–359.

Cognitively challenging talk during shared reading: Effects of parent gender,child gender and relations with story comprehension

Abstract

Keywords

Introduction

Literature review

The current study

Method

Design

Participants

Measures

Observations

Story comprehension

Procedure

Analysis

Results

Descriptive statistics

Research Questions 1 and 2

Research Question 3

Research question 4

Discussion

Limitations and future directions

Conclusion

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

Author Notes

ORCID iD

Appendix

References