Enhancing Engagement With Faculty and Staff to Facilitate Student Success: An Evaluation of a Parent Intervention

Abstract

Many low-income and first-generation students who enroll in college experience less desirable outcomes during their first year. Researchers have increasingly investigated the important role of college knowledge and engagement with faculty and staff for student success. Through a randomized controlled trial intervention, this study leverages the relationship between parents and students to encourage student engagement with faculty and staff during the first year of college. Results of a survey administered to treatment and control students show positive effects of this low-cost, light-touch intervention on parent–student discussions, student attitudes, and intent to persist into the second year of college.

Keywords

educational policy evaluation higher education parents and families social class social stratification experimental research survey research

Introduction

Although the share of first-generation and low-income students enrolled in higher education has increased over time, these students continue to experience less desirable outcomes than their more advantaged peers (Bailey & Dynarski, 2011; Bowen, Chingos, & McPherson, 2009). Ample literature investigates the role of academic preparation and financial supports in socioeconomic inequality (e.g., Bound, Lovenheim, & Turner, 2010; Cabrera, Nora, & Casteneda, 1992; see also a review by Grodsky & Jackson, 2009). These factors, however, do not entirely account for the socioeconomic gap in student experiences and college success (Bowen et al., 2009; Carnevale & Strohl, 2010).

Studies of higher education have identified interaction with faculty and staff as important contributors to postsecondary success, largely because it contributes to student satisfaction with their educational experience (Strauss & Volkwein, 2004; Umbach & Wawrzynski, 2005), improves student grades (Angrist, Lang, & Oreopoulos, 2009), and encourages persistence (MacDonald, Malatest, Assels, Baroud, & Gong, 2009; Tinto, 1993). However, socioeconomically disadvantaged students are less likely to talk with faculty and staff, in part, because they are uncomfortable or unfamiliar with the expectations of the middle-class college context (Collier & Morgan, 2008; Jack, 2016; Yee, 2016). Although prior interventions have tried to provide information directly to students, and encourage their greater engagement with faculty and staff, I shift the focus to consider whether parents from disadvantaged backgrounds can encourage students to engage with institutional agents (i.e., faculty and staff).

Although parents are the most significant source of students’ knowledge of the college system, and communicate frequently with students when they are in college (National Survey of Student Engagement [NSSE], 2007), as 70% of high school completers enroll in college immediately after high school (when they would still have relatively strong ties to parents; U.S. Department of Commerce, Census Bureau, Current Population Survey, 2017), most research on higher education has failed to consider parental influence. Research in K–12 education, however, shows that parents are attentive to education-related information while their children are in high school (Daniel et al., 2009), and that they use this information to make decisions, more so than students themselves (Bettinger, Long, Oreopoulos, & Sanbonmatsu, 2012; Valant & Loeb, 2015). Moreover, parents also use this information to change student behavior (Bergman, 2013; Kraft & Dougherty, 2013; Kraft & Rogers, 2014). The question remains whether parents can help to facilitate student engagement in college.

I developed and evaluated an intervention that targets parents’ conversations with students, enlisting parents to prompt students to contact faculty/staff during college. The sample (N = 617) was evenly split between experimental and control groups. Results based on student surveys indicate that this low-cost, light-touch, parent text-message intervention had an effect on parent–student discussions, how important students believe it is to engage with faculty and staff, and students’ intent to persist into the second year of college. In addition, National Student Clearinghouse (NSC) data indicate there may be effects on actual persistence into students’ second year of college. There is also suggestive evidence that some effects of the intervention may be stronger for students from the least educated families. Overall, these results both document how parents from less advantaged backgrounds can influence students’ college knowledge, thereby encouraging their students to engage with faculty and staff during college and also inform future interventions intended to reduce inequality and promote success among a growing population of students in higher education.

Literature Review

The Role of Parents

Secondary Education

Research on K–12 education demonstrates the integral role parents play in student academic success (Dumais, 2006; Greenman, Bodovski, & Reed, 2011; Jaeger, 2011; Potter & Roksa, 2013). Students learn how to interact with teachers from parents who model interactions with school organizations and staff (Calarco, 2014; Lareau, 2011; Lee & Bowen, 2006). Even parent–student discussions more generally about cultural, social, and political issues are shown to improve student academic performance (Downey, 1995; Jaeger, 2009). Parental influence continues during the transition to college, as input from parents plays a key role in student college choice and enrollment (Deutschlander, 2017; Grodsky & Jones, 2007; Lareau & Weininger, 2008; Myers & Myers, 2012; Roksa & Deutschlander, 2018). In fact, norms and behaviors that parents pass along to their children can give students a persistent advantage throughout their time in the education system. However, the norms and behaviors commonly imparted by middle- and upper-class parents are often supported within the educational system, whereas those practiced by working-class families are less likely to accrue the same advantages (Bourdieu & Passeron, 1977; Calarco, 2014; Lareau, 2011).

Importantly, research suggests that if parents learn new strategies for academic success, they use this information to influence student behavior in school. For example, one intervention with secondary school students sent weekly messages from teachers to parents. Students of parents who received a message indicating what their child could improve in class were less likely to drop out of the summer remediation program (Kraft & Rogers, 2014). Similarly, Kraft and Dougherty (2013) found that frequent teacher–parent phone calls increased student engagement, as measured by homework completion, in-class behavior, and in-class participation during a summer school program. Bergman (2013) also found that sending parents text messages when their child was missing assignments resulted in significant gains in grade point average (GPA), tests scores, and measures of student engagement. The question remains whether parents could be effective in the higher education context.

Postsecondary Education

Once students enter higher education, few studies examine the role of parents in fostering academic success (for recent exceptions, see Auerbach, 2004; Hamilton, 2016; Roksa, Deutschlander, & Whitley, 2016). Despite their absence from the literature, descriptive statistics suggests that parents may be an untapped resource to improve student success in college. Seventy percent of students communicate “very often” with at least one parent (NSSE, 2007), even as much as 13 times per week, with both students and parents initiating contact (Hofer, 2011). Students often report feeling less stressed after communicating with family members (Gemmill & Peterson, 2006; Wolf, Sax, & Harper, 2009). Significantly, parents are an especially important source of support for students who are not in the demographic majority on campus, such as low-income, first-generation, and underrepresented racial/ethnic minorities (Guiffrida, 2006; Melendez & Melendez, 2010).

In addition to serving as a source of emotional support for students, parents also appear to be more attentive to education-related information than college students themselves. For example, one postsecondary institution that sent fliers to both parents and students found parents were more likely than students to remember the information they received (Daniel et al., 2009). This continues a trend documented in secondary education in which parents are more aware of and more likely to use new information than students themselves (Bettinger et al., 2012; Valant & Loeb, 2015). As parents continue to communicate with students frequently, serve as a source of emotional support, and are more likely to retain important college-related information, they may be able to play an important role in helping students navigate postsecondary institutions.

Academic Engagement in College

Descriptive studies of higher education have connected faculty and staff interaction with college success. Students who interact with faculty report significantly higher satisfaction with their overall academic experience (Tinto, 1975; Webber, Krylow, & Zhang, 2013). Moreover, students who report greater satisfaction with faculty interactions also report greater institutional commitment (Strauss & Volkwein, 2004). Both qualitative research on the nature of interactions between faculty and students during college (Deil-Amen, 2011), as well as national quantitative data investigating faculty–student interactions in and outside the classroom (Umbach & Wawrzynski, 2005), suggest that faculty/staff engagement is positively related to student academic achievement, with more engaged students reporting higher levels of learning than unengaged peers. In addition, Webber and colleagues (2013) found that among first-year college students, those who reported one unit increase in interactions with faculty showed a .083 higher cumulative GPA by the end of the year. Moreover, students who report more interactions with faculty during college, inside and outside of the classroom, are more likely to persist in college (Mitchell & Hughes, 2014; Pascarella, Terenzini, & Hible, 1978; Tinto, 1993). Engagement in the college community has long been considered a key to persistence and completion (Tinto, 1975, 1993).

Despite these benefits, many college students do not interact with faculty (Cotton & Wilson, 2006), with even lower rates of faculty and staff interaction among underrepresented racial/ethnic minorities, such as Latinx students (Anaya & Cole, 2001; Rios-Aguilar & Deil-Amen, 2012), low-income students (Stuber, 2011; Yee, 2016), and first-generation students (Collier & Morgan, 2008; Kim & Sax, 2009; Yee, 2016). The discrepancy in faculty/staff engagement between more and less advantaged students is, in part, due to a lack of familiarity with the norms and expectations within the college system (Lareau & Weininger, 2008). Students with limited knowledge of higher education less frequently seek out help while in college, in part, because they do not know what assistance is available and how to access resources when necessary (Lareau, 2011; Stuber, 2011). For example, Lareau (2011) describes one student who could not afford a course textbook. Instead of discussing her problem with the professor, she stopped attending class. Unaware that she should withdraw from the class, she received a failing grade. One relatively minor financial problem turned into a larger problem that needlessly tainted her academic record with a failing grade, not because of academic struggles or directly because of financial problems, but because of lack of knowledge of postsecondary education.

Prior research also suggests that knowledge is not the only reason students do not make connections with faculty or staff on campus. Students may not feel comfortable interacting with faculty (Cotton & Wilson, 2006). Less advantaged college students often display a pattern of independence and hesitancy when interacting with institutional agents (Aries & Seider, 2005; Collier & Morgan, 2008; Jack, 2016; Kim & Sax, 2009; Stuber, 2011; Yee, 2016). Students also avoid faculty because they do not feel entitled to ask for help or may see a request for help as a sign of weakness (Jack, 2016; Lareau, 2015; Yee, 2016). Culture shapes this type of behavior by providing individuals with a tool kit of habits, skills, and styles from which people construct “strategies of action” to solve varying problems—such as those that arise in educational settings (Swidler, 1986). Engaging with faculty or staff can be considered one strategy of action. This is a strategy low-income and first-generation students are less likely to employ than their more advantaged peers (Calarco, 2014; Lareau, 2011). Therefore, one way to change this behavior is to first introduce it as an appropriate strategy of action.

Previous College Student Engagement Interventions

To aid less advantaged students on their journey through higher education, a number of studies have investigated the effects of engagement with faculty, staff, and academic services. Experimental studies that investigate the effect of increased availability of academic services on grades and graduation rates have shown mixed results, mainly because of variation in student engagement with faculty and staff. These studies suggest that informing students of available services is not sufficient to encourage engagement. For example, Angrist and colleagues (2009) found that students who were given access to support services as well as a scholarship increased their grades 3 percentage points relative to students who received nothing. However, students who were made aware of support services and staff help, but not provided with a scholarship, did not improve their grades, suggesting that students may need additional motivation to seek out these types of services. Similarly, Scrivener, Sommo, and Collado (2009) found that the offer of academic services through a College Success Course did not meaningfully affect student academic outcomes unless students were told the course was required.¹ Another intervention that encouraged student interaction with advisors, via email and phone outreach, had no effect on student persistence (Schwebel, Walburn, Klyce, & Jerrolds, 2012). The authors argue that the limited amount of increased interaction with advisors was not substantial enough to affect student outcomes (Schwebel et al., 2012). These college success interventions have been marked by poor engagement with institutional agents.

Students who actually interact with faculty and staff through advising, academic support, and so on show improved postsecondary success, such as higher grades and persistence (Carrell, Kurlaender, & Bhatt, 2018; Clotfelter, Hemelt, & Ladd, 2016; MacDonald et al., 2009; Scrivener & Weiss, 2009). Scrivener and Weiss (2009) showed that students who received US$150 to meet with a guidance counselor for two semesters showed a 4% increase in persistence the semester after the program ended; however, these significant effects dissipated over time. Carrell and colleagues (2018) found that treatment students who received a personalized email from the professor improved their course grade by half a letter grade compared with the control group who received no email. MacDonald et al. (2009) found that at-risk students (who needed a developmental English course; were concerned about integrating into college life; or were uncertain about their academic program/career options) generally do not take advantage of mentoring, tutoring, workshops, or other services that require engagement with college faculty and staff. Specifically, only 14% of students who were encouraged to use these services utilized any type of services during their first semester. However, students who did participate in tutoring, mentoring, or other services for 12 or more hours, had increased persistence into the second semester by 5% (from 61% to 65%) and were twice as likely to graduate (increasing graduation rate from approximately 15%–30%). Notably, these additional services benefited low-income students, students for whom English was a second language, and less academically prepared students the most. Experimental research that does not rely on student initiative, but provides students with additional information through trained coaches who reach out to students, has increased college persistence 5 percentage points (Bettinger & Baker, 2014).

In this study, I consider whether parents can provide the impetus for students to engage with higher education institutions. This research fills two gaps in the field. First, it attempts to improve low levels of student engagement that have contributed to the mixed results in previous college student interventions. Second, despite the significant role that parents play in student cultural capital development, there is a dearth of research on how parents from low-socioeconomic status (SES) backgrounds can support their students during college. To address this, I develop and evaluate an intervention that targets parents’ conversations with students—enlisting parents to prompt students to engage with faculty and staff during college, helping students form relationships that can help them develop cultural capital in college.

Present Study

The Intervention: A Parent Text-Message Program

Although there has been a proliferation of advising programs intended to guide students through college, there are few programs that explicitly recognize and engage parents. The intervention reported here provided parents with an introductory letter at the beginning of students’ first year of college that called upon them, as a vital source of comfort and support, to help their students engage with faculty/staff. Sociological research suggests that whether or not parents intervene is closely tied to what parents believe their role is relative to the education system (i.e., what they are supposed to do in relation to their student’s education) and their knowledge of the education system, which facilitates the fulfillment of their perceived role (Lareau, 2011). The introductory letter suggests to parents that part of their role is to talk to their children about college and presents parents with a straightforward, manageable task. Following the letter, parents received information via text message about specific topics and issues they could discuss with their children throughout the year. A set of three texts, sent biweekly (in either English or Spanish), from August 2016 to May 2017, targeted the content of parents’ conversations with students by identifying particular college engagement strategies. Following the three-message model of York, Loeb, and Doss (2018), each text-set contained: (a) an initial message providing parents with specific information about the importance of certain school-related behaviors; (b) a second message encouraging parents—through a short, specific, and manageable task—to talk with students about the given topic; and (c) a final message reinforcing the value of discussing the suggested topic. Although parents were allowed to respond to text messages, and a college coach was assigned to monitor messages and respond, less than 10% of parents responded to a message. Overall, 15 text-sets were sent biweekly throughout the 2016 to 2017 academic year and covered a range of student engagement strategies. For example, texts described initial faculty and staff outreach, the role of faculty outside of the classroom, how to reach out to faculty and staff for discrete pieces of information, and the importance of building faculty mentoring relationships, among other topics. Text topics were re-introduced throughout the year for reinforcement. An example text set on the topic of professor engagement is italicized below:

Information text: Professors can be intimidating to students, but building close relationships with professors can help students do better in their classes.

Engagement text: Ask your student who their favorite professor is and whether they’ve gone to talk to him or her outside of class.

Encouragement text: By acknowledging that students talk to professors outside of class, you’re helping your student adjust to professor expectations.

Sample and Institutional Partner

A nonprofit organization serving low-income, first-generation, and Latinx students in a southern state, referred to as All Can Achieve (ACA, a pseudonym), executed the parent intervention during the 2016 to 2017 academic year. ACA works with high school students who apply for the program during their junior year of high school. To participate in ACA’s college-access program, students must apply by completing a simple application and essay in the 11th grade of high school.² Students must have a GPA that puts them in the top 60% of their high school class and either (a) qualify for the national school lunch program or (b) be a first-generation student (i.e., neither parent holds a bachelor’s degree, although their siblings may have enrolled in college or completed a degree). The student application must be signed by a parent and includes contact information for one or two parents.

Given the application requirement to participate in ACA, the families participating in this study are likely not representative of the nationwide population of low-income and/or first-generation families because either parent or student initiative was necessary to join the high school program. Also, although ACA does not target specific racial/ethnic minority groups, their geographic location means that many of the participants are Latinx students. Although the number of Latinx students in this study may be disproportionate to the nation as a whole, Latinx students are the largest and fastest growing minority group in the United States (Fry, 2011). The share of Latinx high school graduates enrolled in college immediately after high school reached 49% in 2012, surpassing that of Whites (Fry & Taylor, 2013). Importantly, the proportion of Latinx students who are first-generation students is also higher than any other group—approximately 50% (National Center for Education Statistics [NCES], 2010). Moreover, this population of college students has particularly low college completion rates, with only 36% of first-time, full-time Latinx students earning a degree within 6 years, compared with 49% of Whites (NCES, 2011).

Table 1 shows that approximately 70% of students in the sample are Latinx, whereas 8% are White and approximately 18% are African American. Also, nearly one quarter reported that their parents’ preferred language was Spanish. Most of the students come from less-educated backgrounds—over 65% would be the first in their family to earn a bachelor’s degree. Nearly 75% of students qualified for free/reduced price lunch (FRPL) in high school. As expected, based on ACA’s application requirements, these students reported an average GPA of 3.0 (the majority falling between 2.5 and 3.5), with over 90% in the top 60% of their graduating class in their junior year of high school. Most students are female (nearly 65%).

Table 1

Experimental Sample Summary Statistics

Variables	M	SD
Female	0.64
Race/ethnicity
Latinx	0.69
White	0.08
African American	0.18
Other	0.05
High School
Junior GPA	3.16	0.53
Top 60% of class	0.91
ACA attendance^a	0.53	0.23
Family background
First-generation	0.66
Free/reduced price lunch	0.74
Language: parents prefer English	0.29
Language: parents prefer Spanish	0.26
College type
2-year college	0.31
4-year college	0.46
College not specified	0.21
N	617

Note. M = mean; SD = standard deviation; GPA = grade point average; ACA = All Can Achieve.

Rate of attendance at ACA’s after-school college-access classes.

The experimental sample consists of 617 families from the ACA class of 2016 cohort. In mid-August of 2016, ACA families were randomly assigned to treatment or control conditions (control, n = 309; treatment, n = 308). Subsequently, ACA reached out to and implemented the intervention with 308 treatment group parents and guardians of the 2016 cohort. Initial letters went out to 306 families (addresses were unavailable for 2 families). Text messages were sent to 256 families—37 families did not have parent cell information and 15 students did not have a parent on file. Accurate contact information is often difficult to collect among less advantaged groups; therefore, this amount of missing data is not surprising. For 57% of the families receiving text messages, ACA had contact information for and sent messages to two parents. College-access coaches update student and parent contact information throughout students’ tenure in the program, which likely contributes to the large proportion of parents with available contact information. Over the course of the academic year, 44 parents opted out of text messages. Of these, approximately half still had a spouse receiving messages. As a result, only 17 out of 256 families receiving text messages left the study completely—an opt-out rate of less than 7%. Of the 308 families in the sample, 235 received the full intervention (letters and text messages) throughout the year. The analytic sample is based on a subset—students surveyed in spring 2017—of the entire experimental sample (617) regardless of whether parents received text messages or opted out of the program.

Analytical Strategy

To evaluate the impact of assignment to the parent intervention, I randomly assigned students, and their families, to treatment and control conditions and tested for baseline equivalence on student characteristics before the start of the intervention. Table 2 presents the results of regression analyses predicting student baseline covariates with an indicator reflecting assignment to treatment. There is no statistically significant imbalance on observable characteristics between treatment and control groups.

Table 2

Treatment-Control Group Balance Tests

Variables	Control M	Treatment M
Female	0.645	0.625
Race/ethnicity
Latinx	0.691	0.686
White	0.084	0.095
African American	0.158	0.045
Other	0.067	−0.027
High School
Junior GPA	3.174	3.143
Top 60% of class	0.917	0.906
ACA attendance^a	0.522	0.547
Family background
First-generation	0.651	0.676
Free/reduced price lunch	0.730	0.753
Language: parents prefer English	0.317	0.267
Language: parents prefer Spanish	0.239	0.205
College type
2-year college	0.285	0.325
4-year college	0.470	0.452
College not specified	0.228	0.188
N	309	308

Note. M = mean; GPA = grade point average; ACA = All Can Achieve.

Rate of attendance at ACA’s after-school college-access classes.

Statistically significant differences between control and treatment group means are indicated by the following symbols: ^†p < .10. *p < .05. **p < .01. ***p < .001.

To investigate the effects of the parent intervention, this study employs survey data from both treatment and control students collected during their first and second semesters in college (collected in November 2016 and March 2017, respectively).

Survey respondents answered questions that investigate whether the intervention impacted the content of parent–student conversations, as well as students’ attitudes and behaviors related to faculty and staff engagement. Students were asked how often they communicate with their parents and how often they discuss the following college topics: professors, academic advisors, meetings with advisors, relationships with professors outside of class, course assignments, and so on. Parental discussions were measured on a 5-point Likert-type scale from never to always. Students also reported their attitudes toward engagement, measured on a Likert-type scale indicating how important it is to talk with and meet with faculty and teaching assistants during college. Students also reported their actual engagement with faculty and staff, measured through how many times students talked to professors, academic advisors, visited the writing center, and so on.

Student intent to persist is measured with a question indicating how likely students are to return to college in the fall 2017. Closely related to persistence, student goal and institutional commitment are captured with measures that ask whether students agree or disagree with the statement that they are pleased with their decision to attend college and, in particular, their current institution. In addition, student enrollment in the second year of college (Fall Semester, 2017) was captured through NSC data.

As all survey measures consist of Likert-type scales, I use ordinal logistic regression to estimate the intent-to-treat (ITT) effects of the intervention on student outcomes.³ ITT analyses compare mean outcomes of groups as randomized. This type of analysis is appropriate given that in most cases it is not possible to know whether the parents actually received, read or discussed the text messages with their children. The analytic model is estimated in Stata and therefore specified as follows (Liu, 2009):

\begin{array}{l} l n (Y_{j}^{'}) = l o g i t [π (x)] = l n (\frac{π_{j} (x)}{1 - π_{j} (x)}) = \\ α_{j} + (- β_{1} T R E A T M E N T_{i} - γ X_{i} - ε_{i}) \end{array}

where π_i(x) = π (Y ≤ j| x₁, x₂, …, x_p) is the probability of being at or below category j, given a set of predictors; j =1, 2, … J − 1; α_j are the cutpoints; TREATMENT_i is a binary indicator for whether students’ parents have been randomly assigned to participate in the parent intervention or to the control group; X_i is a vector of nine student-level sociodemographic baseline covariates collected from the ACA application, for both treatment and control students; and ε_i is a residual error term. The sociodemographic controls include the following: indicator variables for female students, Latinx students, and 2-year college attendance, as well as an indicator for FRPL status, a categorical measure of parental educational level, an indicator for English spoken at home, a continuous measure of students’ high school GPA, an indicator for high school attended, and a continuous measure of students’ rate of attendance at ACA’s after-school college-access program. In this model, β₁ provides an unbiased causal estimate of the impact of the offer to participate in the intervention on student outcomes.⁴ In addition, to estimate the effect of treatment on student persistence into the Fall 2017 semester, I use a linear regression framework for ease of interpretation. Supplemental analysis with logistic regression shows a similar pattern.

In addition to estimating main effects, I also test for heterogeneous effects to determine whether or not the parent intervention had a differential effect for particular families. Given that parental education level is linked to student engagement in college (Collier & Morgan, 2008; Yee, 2016), and students in the sample come from varying educational backgrounds, students may experience differential effects based on family educational background. Given concerns of power, I do not test for the possibility of other heterogeneous treatment effects. To test for the effect of parental education, I add interactions terms to the fully specified models. Given the statistical power of this study, I combine parental education levels into two categories: (a) some college, AA degree, BA degree or more; (b) high school diploma or less.

Finally, I include supplementary analyses that account for multiple hypothesis testing with the Benjamini–Hochberg (1995) method.⁵ This method adjusts for multiple comparisons by controlling the false discovery rate (instead of the familywise error rate). These results are included in Supplemental Appendix and mentioned below where relevant.

Attrition Analyses

This study employs survey data from both treatment and control students collected during their first and second semesters in college (collected in November 2016 and March 2017, respectively). With a study like this, which relies on survey data for analysis, the difference in the share of students included in the experimental sample and survey samples represents attrition from the experiment. The survey yielded a response rate of 51%,⁶ resulting in a survey sample of approximately 300 students, compared with the experimental sample of 617. Attrition can lead to biased estimates of impact if the types of treatment group students who attrited (did not respond to the survey) are systematically different than the type of control group students who attrited in a way that is related to survey measures outcomes.

To analyze attrition in the student survey data, I test whether survey response differs by treatment status. Specifically, I regress a binary variable that equals 1 if a student responded to the survey and 0 if not on treatment status. I find no evidence that the rate of attrition (survey response) differs between the treatment and control groups (Table 3; see Supplemental Appendix for further attrition analyses).

Table 3

Survey Attrition, Samples, and Descriptive Statistics

	Nonrespondents	Respondents
Percent of students in treatment group	0.52	0.45
			Respondents
Variables			Control	Treatment
Female	0.59	0.68*	0.68	0.68
Race/ethnicity
Latinx	0.71	0.67	0.68	0.66
White	0.08	0.08	0.09	0.06
African American	0.18	0.18	0.16	0.21
Other	0.04	0.07^†	0.07	0.07
High school
Junior GPA	3.01	3.29***	3.25	3.33
Top 60% of class	0.89	0.94^†	0.93	0.94
ACA attendance^a	0.50	0.57***	0.54	0.61**
Family background
First-generation	0.64	0.69	0.69	0.68
Parental education^b
Some college, AA, BA or more			0.35	0.37
High school graduate or less			0.54	0.47
Missing			0.11	0.17
Free/reduced price lunch	0.72	0.77	0.77	0.76
Language: parents prefer English	0.26	0.33^†	0.35	0.30
Language: parents prefer Spanish	0.23	0.28	0.26	0.30
College type
2-year college	0.35	0.27^†	0.29	0.28
4-year college	0.36	0.52***	0.54	0.58
College not specified	0.29	0.21***	0.17	0.13
N	318	299	160	139
Response rate		50.94%

Note. Column 1 shows the proportion of survey nonrespondents in the treatment group. Column 2 shows the proportion of survey respondents in the treatment group and indicates if this difference is statistically significant. (This test, which acts as a test of differential attrition, was calculated by regressing a binary variable equal to one if a student responded to the survey on treatment status, to determine if treatment assignment could predict survey response.) Individual regression analyses omit students who are missing data. GPA = grade point average; ACA = All Can Achieve; AA = associate’s degree; BA= bachelor’s degree.

Rate of attendance at ACA’s after-school college-access classes.

Data available for survey sample only because ACA only collects data that indicates whether parents have a bachelor’s degree.

Statistical significance levels: ^†p < .10. *p < .05. **p < .01. ***p < .001.

There are two reasons the survey response rate and potential attrition should not affect treatment effects. First, the intervention is not likely to affect who decides to complete the survey because most students were unaware that their parents received messages from ACA. Interviews with a subsample of treatment parents suggest that many parents worked the topics of text messages into regular conversations and did not tell students that these were suggested by ACA. Correspondingly, treatment assignment is a poor predictor of whether or not students report that their parents received messages from ACA during the academic year. As most students are unaware of their treatment status, they are unlikely to feel more or less compelled to complete the survey. Second, if, as Supplemental Appendix Table 1 (in the online version of the journal) suggests, higher academic achievers and students with higher ACA attendance are more likely to complete the survey, and high school academic achievement and ACA attendance are positively related to academic success in college, then any effects would be understated. As poorer academic performers are less likely to engage with faculty and staff during college (Carini, Kuh, & Klein, 2006), they have the greatest potential for improvement and could show the greatest treatment effects. In addition, students who are less engaged with ACA during high school are probably less likely to engage with faculty and staff in college, because students often continue practices from high school into college (Cole, Kennedy, & Ben-Avie, 2009; Jack, 2016). Therefore, if students with low engagement rates are less likely to complete the survey, then the survey reported control group mean of engagement is likely higher than it is among the entire ACA population. As a result, the sample from the survey is not likely to bias results in a way that would artificially inflate treatment estimates, but rather depress them.

Survey respondents differ from nonrespondents on important demographic characteristics, which limits generalizability of the survey sample to the overall experimental sample. As Table 3 indicates, compared with nonrespondents, survey respondents are disproportionately female, younger, higher academic achievers, and 4-year college attendees. Across the experimental sample and survey sample, the proportion of students from various racial/ethnic groups is similar.

Table 3 also shows that survey respondents were split between treatment and control group students, with 53.5% of the survey sample from the control group and 46.5% from the treatment group. Also, the characteristics of the respondents are similar across treatment and control groups, which supports the internal validity of survey measures. There is only one statistically significant difference between treatment and control responders on baseline characteristics—treatment responders had higher rates of ACA attendance in high schools (attending 60.9% of classes instead of 53.7% in the control group).⁷

Findings

Analyses of a Fall 2016 survey suggest that although parent–student discussions of academic topics increased for treatment students, student engagement with faculty and staff,⁸ and persistence were not affected. This is not surprising given the limited duration of the intervention at the point of the fall survey distribution. At the time of the fall survey, the intervention had been in the field for 2 to 3 months. Parents had received an introductory letter and between four and six text-sets. Because of these limited effects in the fall, the results presented here focus on the Spring 2017 survey, which was administered after the intervention was in the field for more than 6 months. Fall 2016 findings available upon request from the author.

To test the fidelity of implementation of the intervention, I first investigate parental discussions with students. Table 4 reports the difference between treatment and control group students’ survey responses. Models accounting for multiple hypothesis testing are shown in Supplemental Appendix Table 4 (in the online version of the journal). Where results differ, I suggest results be interpreted with caution. As expected, students report that on average they communicate with parents multiple times a day, with most students communicating with their parents between 2 and 5 times a day. Survey responses show that students in the treatment group are not more likely to talk to their parents than control students, but rather their topics of conversation differ. On average, students in the control group reported that they rarely discussed academic services, their academic advisor, their professors,⁹ or relationships with their professors outside of class. Treatment group students reported higher odds of having these conversations with parents. Some coefficients are reduced, whereas others increase as academic and sociodemographic covariates are added to the model (see Model 2), but statistical significance levels persist. The coefficients in Model 2 indicate that treatment students show 1.48 greater odds of reporting that they always discuss academic services compared with the other categories of conversation frequency (including never to most of the time). In addition, treatment students show 1.63 and 1.56 greater odds of always having conversations related to academic advisors, and 1.87 and 1.72 greater odds of always having conversations about professors and relationships with professors outside of class compared with other categories of conversation frequency.¹⁰ Overall, this intervention generated small change¹¹ in the awareness of important elements of postsecondary institutions among parents and students.

Table 4

Effect of Treatment Assignment on Parent–Student Discussions, Student Attitudes, and Behaviors

Parent–student discussions	Control M	Treatment effects
Parent–student discussions	Control M	Model 1	Model 2
How frequently do you communicate with your parents via text, phone, email, etc.? Scale from 1 to 8: Not at all (1), A few times a month (2), Once a week (3), A few times a week (4), Once a day (5), 2–3 times a day (6), 4–5 times a day (7), and Every 2 hr or more (8).	5.450	1.050	1.063
How often do you talk about: Scale from 1 to 5: Never (1), Rarely (2), About half the time (3), Most of the time (4), and Always (5).
Academic services (e.g., tutoring or writing center)	2.150	1.481^†	1.477^†
Your academic advisor	2.000	1.566*	1.626*
Meetings with your academic advisor	2.106	1.519*	1.559*
Your classes	3.388	1.201	1.242
Studying/preparing for class	3.200	1.282	1.229
Class assignments	2.925	1.312	1.221
Your professors	2.294	1.793**	1.870**
Your relationships with professors outside of class	1.918	1.626*	1.720*
Student attitudes
While in college, how important is it that students do the following? Scale from 1 to 5: Not at all important (1), Not very important (2), Somewhat important (3), Important (4), and Very important (5).
Talk with professors/TAs during class.	3.737	1.483^†	1.392
Talk with professors/TAs about academic performance in class.	3.869	2.014**	2.229**
Talk with professors/TAs one-on-one outside of class.	3.812	1.588*	1.523^†
Go to a professors/TAs office hours.	3.912	1.683*	1.555^†
Ask professors, TAs, or staff for advice.	4.006	1.599*	1.563^†
Develop a relationship with a professors, TA, or staff member.	3.906	1.397^†	1.277
Student behaviors
In college this semester (Spring 2017), how many times have you: 0 (0), 1 (1), 2–3 times (2), 4–5 times (3), and 6 or more (4).
Talked to a professor or TA outside of class?	1.887	1.327	1.350
Talked to an academic advisor?	1.988	0.839	0.829
Talked to other staff?	1.650	1.378^†	1.360
Visited the academic support center?	1.346	1.179	1.169
Visited the writing center?	0.912	0.727	0.741
Student intent to persist
I will attend college next Fall 2017^a	4.631	2.618**	2.835**
I am pleased with my decision to go to college^b	5.300	1.325	1.302
I am pleased with my decision to go to this college^b	5.006	0.981	0.942
Model inclusions
Controls			Yes
Sample size	299	299	299

Note. This table reports odds ratios coefficients from separate ordinal logistic regressions estimated for each survey outcome of interest. Controls: female, indicator for Latinx, parental education level, free/reduced price lunch status, language spoken at home, student high school GPA, student attendance in the All Can Achieve HS program, HS, and indicator for 2-year college attendance. M = mean; TA = teaching assistant; GPA = grade point average; HS = high school.

Scale from 1 to 5: Very unlikely (1), Somewhat unlikely (2), Undecided (3), Somewhat likely (4), and Very likely (5).

Scale from 1 to 6: Strongly disagree (1), Disagree (2), Somewhat disagree (3), Somewhat agree (4), Agree (5), and Strongly agree (6).

Statistical significance levels: ^†p < .10. *p < .05. **p < .01. ***p < .001.

To investigate whether student attitudes toward faculty engagement changed, students were asked about the importance of various types of engagement. The analyses presented in Table 4 show that on average control students reported it is important to reach out to faculty and teaching assistants. Despite a high degree of agreement with the importance of engagement among the control group, Model 2 shows that treatment students still display higher odds of agreement than control students that talking with professors and teaching assistants about academic performance in class is important (odds of a strongly agree response increase by a factor of 2.23 for treatment students). In addition, odds ratios that represent how important students believe it is to talk with faculty and teaching assistants outside of class, during office hours, and ask them for advice show students’ odds of reporting they strongly agree instead of compared with other categories (strongly disagree–agree), are approximately 1.5 times greater for treatment students than control students.

The intervention also appears to have had no measureable effect on student behavior. Control group students reported that they interacted with professors or teaching assistants less than 2 to 3 times during the spring semester. Statistically insignificant coefficients suggest that treatment group students may show 1.3 times greater odds of reporting higher interactions with faculty, teachings assistants, or other staff. Surprisingly, odds ratios below 1 suggest that treatment may reduce students’ odds of visits to the writing center or academic advisors. Additional discussions with professors, teaching assistants, and other staff may have reduced the need for visits to the writing center.

Most importantly, treatment students are significantly more likely to report that they plan to attend college in Fall 2017 (their second year of college). This is especially noteworthy given that the control group reports a high likelihood of intent to persist—4.6 on a 5-point scale (between somewhat likely and very likely). Therefore, any significant increase above and beyond this high intent to persist is remarkable. Treatment students experienced a dramatic 2.84 times greater odds of stating they are very likely to persist into their second year of college compared with control students. In addition, analyses of NSC data of the survey sample, shown in Table 5, indicate that the intervention may have resulted in a 5 percentage-point increase in persistence into the second year of college. However, NSC data suffer from differential attrition and the point estimate is statistically insignificant (p = .224).¹² Therefore, these results should be interpreted with caution. These statistically insignificant findings could result from a number of sources, the incomplete NSC data, small sample size, or limited program duration and intensity.

Table 5

Effect of Treatment Assignment on Persistence Among Wave II Survey Respondents

	Control M	Treatment effects
	Control M	Model 1	Model 2
Enrollment in second year of college	0.86	0.056 (0.036)	0.053 (0.039)
Enrollment in second year of college	0.86	p = .122	p = .177
Model inclusions
Controls			Yes
Sample size		299	299

Note. Standard errors in parentheses. Controls: female, indicator for Latinx, parental education level, free/reduced price lunch status, language spoken at home, student high school GPA, student attendance in the All Can Achieve HS program, HS, and indicator for 2-year college attendance. M = mean; GPA = grade point average; HS = high school.

Heterogeneous Effects

Table 6 examines whether there are differential effects of the intervention for students whose parents had some college experience or a degree versus students whose parents had a high school diploma or less. In other words, Table 6 reports the effect of the parent program on students from less-educated families above and beyond the effect of the parent program on students from more-educated families. These results should be interpreted with caution as the significance level of the p-values does not pass the Benjamini–Hochberg test for multiple comparisons. One statistically significant interaction among parent–student discussions indicates that students from less-educated backgrounds show increased odds of conversations with parents related to studying and preparing for class relative to students from more-educated backgrounds in the treatment group. As well, although statistically insignificant, the size and direction of additional interaction coefficients suggest that less-educated parents in the treatment group may have increased odds of conversations related to class assignments and professors relative to more-educated parents in the treatment group, and may have lower odds of discussing relationships with professors relative to more-educated parents. Supplemental analyses with separate regression analyses by parental education level (see Supplemental Appendix Table 3 in the online version of the journal) show different control group means among the two student groups. Moreover, statistically significant t-tests reveal that among control group students, more-educated families discuss studying/preparing for class and professors more frequently than less-educated families. However, treatment students from less-educated families report that they discuss these topics as frequently as treatment students in more-educated families.¹³ In these instances, the intervention appears to reduce the gap in academic discussions typically seen between families of different parental education levels. The odds ratios below 1 in Table 6, although not statistically significant, suggest that treatment students from less-educated backgrounds may have lower odds of conversations with parents about academic advisors or relationships with professors than treatment students from more-educated backgrounds.¹⁴

Table 6

Heterogeneity of Treatment Effects by Parental Education—Interaction Effects

Treatment × Parental Education^a
Parent–student discussions	HS diploma or less
How frequently do you communicate with your parents via text, phone, email, etc.? Scale 1–8: Not at all (1), A few times a month (2), Once a week (3), A few times a week (4), Once a day (5), 2–3 times a day (6), 4–5 times a day (7), and Every 2 hr or more (8).	0.690
How often do you talk about: Scale 1–5: Never (1), Rarely (2), About half the time (3), Most of the time (4), and Always (5).
Academic services (e.g., tutoring or writing center)	0.781
Your academic advisor	0.614
Meetings with your academic advisor	0.720
Your classes	1.369
Studying/preparing for class	2.245^†
Class assignments	1.535
Your professors	1.627
Your relationships with professors outside of class	0.600
Student attitudes
While in college, how important is it that students do the following? Scale 1–5: Not at all important (1), Not very important (2), Somewhat important (3), Important (4), and Very important (5).
Talk with professors/TAs during class.	2.737*
Talk with professors/TAs about academic performance in class.	1.397
Talk with professors/TAs one-on-one outside of class.	1.588
Go to a professors/TAs office hours.	1.733
Ask professors, TAs, or staff for advice.	2.134^†
Develop a relationship with a professors, TA, or staff member.	1.024
Student behaviors
In college this semester (Spring 2017), how many times have you: 0 (0), 1 (1), 2–3 times (2), 4–5 times (3), and 6 or more (4).
Talked to a professor or TA outside of class?	0.940
Talked to an academic advisor?	0.447^†
Talked to other staff?	1.012
Visited the academic support center?	0.950
Visited the writing center?	0.473
Student intent to persist
I will attend college next Fall 2017^b	3.459^†
I am pleased with decision to go to college^c	1.858
I am pleased with decision to go to this college^c	1.821
Model inclusions
Controls	Yes
Sample size	257

Note. This table reports odds ratios coefficients from separate ordinal logistic regressions estimated for each survey outcome of interest. Controls: female, indicator for Latinx, free/reduced price lunch status, language spoken at home, student high school GPA, student attendance in the All Can Achieve HS program, HS, and indicator for 2-year college attendance. HS = high school; TA = teaching assistant; GPA = grade point average.

Reference: Some college, AA, BA, or more.

Scale 1 to 5: Very unlikely (1), Somewhat unlikely (2), Undecided (3), Somewhat likely (4), and Very likely (5).

Scale 1 to 6: Strongly disagree (1), Disagree (2), Somewhat disagree (3), Somewhat agree (4), Agree (5), and Strongly agree (6).

Statistical significance levels: ^†p < .10. *p < .05. **p < .01. ***p < .001.

Although the interaction effects of parent–student conversations are mixed, the parent intervention appears to have had a significantly stronger effect on students’ attitudes from less-educated backgrounds. More specifically, treatment students from less-educated families experience 2.74 times greater odds of reporting they believe talking with professors and teaching assistants during class is very important than treatment students from more-educated backgrounds. In addition, treatment students from less-educated backgrounds show 2.13 greater odds of reporting that asking professors and teaching assistants for advice is very important.¹⁵ In addition, although other interaction effects related to students attitudes are not statistically significant, the coefficients suggest that students from less-educated backgrounds may also experience greater odds of reporting that other types of engagement is very important, specifically talking to professors or teaching assistants during class, one-on-one outside of class, or during office hours (between 1.4 and 1.7 times greater odds).

The treatment appeared to have less of an effect on behavior for students from less-educated backgrounds than students from more-educated backgrounds. Students from less-educated backgrounds had .45 times lower odds of reporting frequent interactions with academic advisors as a result of the intervention than treatment students from more-educated backgrounds. In addition, another odds ratio below 1 related to visiting the writing center suggests that treatment was more effective at increasing actual engagement among students from more-educated backgrounds. Analyses in Supplemental Appendix Table 3 (in the online version of the journal) suggest that treatment may have decreased engagement with the writing center for students from less-educated backgrounds, as these students show lower odds of visiting the writing center than their control group peers (a similar, statistically insignificant, pattern is seen among engagement with academic advisors).

Finally, the treatment had a greater effect on student intent to persist among students whose parents possess a high school diploma or less, than students from more-educated backgrounds. Students from less-educated backgrounds experienced 3.46 times greater odds of intent to persist than students from more-educated backgrounds. In addition, strong, yet statistically insignificant coefficients suggest that students from less-educate backgrounds show greater odds of strongly agreeing with goal commitment and institutional commitment statements above and beyond that of treatment students from more-educated backgrounds. Supplemental Appendix Table 3 (in the online version of the journal) shows that this is partly due to the lower means among the control group students from less-educated backgrounds and null-treatment effects among students from more-educated backgrounds. Control group students from less-educated backgrounds were less likely to indicate that they were pleased with their college choice, with their choice to attend college, or planned to persist into their second year of college than control group students from more-educated backgrounds. Overall, these interaction analyses suggest that the intervention had the most significant effect on student attitudes related to engagement and persistence among treatment students from less-educated families.

Discussion

Although the share of first-generation and low-income students enrolled in higher education has increased over time, they continue to graduate at lower rates and are more likely to leave after the first year than their more advantaged peers (Bowen et al., 2009). Studies of higher education have thus increasingly investigated the knowledge and practices that facilitate successful interaction with social institutions, specifically engagement with faculty and staff—as a contributor to socioeconomic inequality (Lareau, 2011). Importantly, students with less college knowledge, often the first in their family to go to college, are less likely to interact with faculty and staff (Jack, 2016). To encourage socioeconomically disadvantaged students to engage with faculty and staff during college, I developed and evaluated an intervention that targets parents’ conversations with students, enlisting parents to prompt and remind students to contact faculty and staff.

The intervention significantly increased the number of conversations between parents and students about college throughout the year, changed student attitudes, and positively influenced their intent to persist into their second year of college (with suggestive evidence that actual persistence may have been impacted). This suggests not only that parent–student conversations during college matter but also that they can have a causal effect on student attitudes. This is remarkable given the light-touch character of the intervention, which was administered through letters and text messages to parents during students’ first year of college. The results of this parent intervention have important implications for practitioners, researchers, and policymakers interested in student engagement and reducing inequality in college persistence. First, the long-term assessment of the intervention, provided by the spring 2017 survey, was crucial to understanding the effect this intervention had on first-year college students. The lack of significant changes in student attitudes and persistence in the fall of 2016 and significant changes in the spring of 2017, indicate that this type of intervention may need more than one semester to make an impact.

An even longer assessment of this type of intervention may be crucial to detect changes in student behavior.¹⁶ Research from the social and health sciences show that interventions typically have smaller effects on individual behavior than attitudes. In a meta-analysis of health research, Sheeran and colleagues (2016) found that, on average, interventions designed to change patient attitudes had a medium-sized effect on an individual’s intentions, but a smaller effect on behavior. In a college intervention designed to increase the persistence of first-generation students, psychologists found that the effect on student attitudes was greater than the effect on student behaviors (Stephens, Hamedani, & Destin, 2014). Researchers theorize that this pattern is partly due to the difficulty of changing behavior based on social norms (Miller & Prentice, 2016) and because habits often persist after intention changes (Gardner, de Bruijn, & Lally, 2011). As a result, changes to behavior may be more difficult to achieve than changes to attitudes.

In addition, the significant impact on attitudes and the limited impact on student behavior may be due to two empirical characteristics of the study. First, this study may not be powered to detect a small effect in behavior given the sample size. Second, the measures of behavior do not cover the entire semester. Timing of the survey is significant given that many student–faculty interactions may happen toward the end of the semester as students start getting feedback on their performance in class and prepare for final papers and exams. Sociologists argue that individuals often need a problem that can be solved with a new strategy of action before individuals are likely to employ it (Joas, 1996; Sewell, 1992; Swidler, 1986). Thus, as the survey was administered in the middle of the semester and does not capture the latter half of the semester—when more pressing problems may present themselves—students may have indeed changed their behavior without the measures here capturing it.

The second implication from this research is that analyses of heterogeneous effects suggest that there may be some variations among students from different educational backgrounds. The positive effects on student attitudes (as measured through how important they believed engagement to be as well as their intention to persist) are more pronounced among ACA students whose parents do not have any experience with college. As students whose parents have less experience with higher education may benefit more, it may be important to target these groups in future interventions.

Especially noteworthy for future intervention research, this study suggests that parents can influence college students’ attitudes despite their own limited experience with college. Although previous research suggested that parents could change high school students’ behavior (Bergman, 2013; Kraft & Dougherty, 2013; Kraft & Rogers, 2014), it was unclear to what degree parents could influence college students. The magnitude of the effect on persistence is similar to other intervention research in higher education, which show 3 to 5 percentage-point increases in persistence (Clotfelter et al., 2016; MacDonald et al., 2009; Scrivener & Weiss, 2009). In fact, a potential 5 percentage-point increase would be noteworthy given that Clotfelter and colleagues (2016), as well as Scrivener and colleagues (2009), investigate much more comprehensive interventions.

The effects reported here are surprising given the low-cost, light-touch nature of this intervention. The cost of a year-long parent intervention is approximately US$9,250, which breaks down to nearly $30 per family. This per-family rate would decrease for large-scale interventions, as texting costs would decline for long-term and large-sample programs. This type of intervention could be paired with interventions more common in the higher education context—in which institutions increase outreach to students to raise awareness of available academic services or provide additional academic resources to target groups. Additional ways to amplify the effects shown here could be to extend the duration of the intervention or add a concrete component to the intervention. For example, an intervention that combines parental encouragement with models for how students might engage with faculty and staff via email or in-person meetings would likely be more powerful. Given that this intervention involves no sharing of student data with parents, only tips about how to engage with faculty and staff, there are no constraints on institutions of higher education (such as Family Educational Rights and Privacy Act [FERPA]) from employing an intervention like this to increase persistence among students.

Supplemental Material

DS_10.3102_0162373719845653 – Supplemental material for Enhancing Engagement With Faculty and Staff to Facilitate Student Success: An Evaluation of a Parent Intervention

Supplemental material, DS_10.3102_0162373719845653 for Enhancing Engagement With Faculty and Staff to Facilitate Student Success: An Evaluation of a Parent Intervention by Denise Deutschlander in Educational Evaluation and Policy Analysis

Footnotes

Acknowledgements

The author thanks Josipa Roksa for her vital feedback and support of this research. The author also thanks Sarah Mosseri, Ben Castleman, Dan Player, Paul Kingston, and EEPA reviewers for their insightful comments.

Declaration of Conflicting Interests

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

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research is supported by the NSF-Social and Economic Sciences Doctoral Dissertation Research program (grant 1602743) and the Institute of Education Sciences, U.S. Department of Education, through grant #R305B090002 to the University of Virginia. The opinions expressed are those of the author and do not represent views of the institute or the U.S. Department of Education. Funding was also provided by the University of Virginia Quantitative Collaborative through the Bynum Grant.

Notes

ORCID iD

D. Deutschlander

Author

Denise Deutschlander is a senior researcher in the Center for Research Evaluation and Assessment at Education Northwest. Her research explores the impact of policies and programs designed to improve postsecondary access and success for underrepresented groups, with particular focus on the relationship between family context and student success.

References

Anaya

Cole

(2001). Latina/o student achievement: Exploring the influence of student–faculty interactions on college grades. Journal of College Student Development, 42, 3–14.

Angrist

Lang

Oreopoulos

(2009). Incentives and services for college achievement: Evidence from a randomized trial. American Economic Journal: Applied Economics, 1, 136–163.

Auerbach

(2004). Engaging Latino parents in supporting college pathways: Lessons from a college access program. Journal of Hispanic Higher Education, 3, 125–145.

Aries

Seider

(2005). The interactive relationship between class identity and the college experience: The case of lower income students. Qualitative Sociology, 28, 419–443.

Bailey

Dynarski

(2011). Inequality in postsecondary education In Duncan

G. J.

Murnane

R. J.

(Eds.), Whither opportunity: Rising inequality, schools, and children’s life chances (pp. 117-132). New York, NY: Russell Sage Foundation.

Benjamini

Hochberg

(1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B Methodological, 57, 289–300.

Bergman

(2013). Bergman, P. 2015. Parent-child information frictions and human capital investment: Evidence form a field experiment (CESifo Working Paper Series No. 5391). New York, NY: Columbia University Teachers College.

Bettinger

Baker

(2014). The effects of student coaching: An evaluation of a randomized experiment in student advising. Educational Evaluation and Policy Analysis, 36, 3–19.

Bettinger

Long

Oreopoulos

Sanbonmatsu

(2012). The role of application assistance and information in college decisions: Results from the H&R block FAFSA experiment. The Quarterly Journal of Economics, 127, 1205–1242.

10.

Bound

Lovenheim

Turner

(2010). Why have college completion rates declined? An analysis of changing student preparation and collegiate resources. American Economic Journal: Applied Economics, 2, 129–157.

11.

Bourdieu

Passeron

J. C.

(1977). Reproduction in education: Society and culture. London, England: SAGE.

12.

Bowen

Chingos

McPherson

(2009). Crossing the finish line: Completing College at America’s public universities. Princeton, NJ: Princeton University Press.

13.

Cabrera

Nora

Casteneda

(1992). The role of finances in the persistence process: A structural model. Research in Higher Education, 33, 571–593.

14.

Calarco

(2014). Coached for the classroom: Parent’s cultural transmission and children’s reproduction of educational inequalities. American Sociological Review, 79, 1015–1037.

15.

Carini

Kuh

Klein

(2006). Student engagement and student learning: Testing the linkages. Research in Higher Education, 47(1), 1–32.

16.

Carnevale

Strohl

(2010). How increasing college access is increasing inequality, and what to do about it In Kahlenberg

(Ed.), Rewarding strivers: Helping low-income students succeed in college (pp. 1–23). New York, NY: Century Foundation Books.

17.

Carrell

Kurlaender

Bhatt

(2018). My professor cares: Experimental evidence on the role of faculty engagement (Working paper). March presentation at Oregon State University, School of Public Policy. Retrieved from http://faculty.econ.ucdavis.edu/faculty/scarrell/

18.

Chen

Cohen

Chen

(2010). How big is a big odds ratio? Interpreting the magnitudes of odds ratios in epidemiological studies. Communication in Statistics—Simulation and Computation, 39, 860–864.

19.

Clotfelter

C. T.

Hemelt

S. W.

Ladd

H. F.

(2016). Multifaceted aid for low-income students and college outcomes: Evidence from North Carolina. Economic Inquiry, 56, 278–303. Retrieved from http://onlinelibrary.wiley.com/doi/10.1111/ecin.12486/abstract

20.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.

21.

Cole

Kennedy

Ben-Avie

(2009). The role of precollege data in assessing and understanding student engagement in college. New Directions for Institutional Research, 141, 55–69.

22.

Collier

P. D.

Morgan

(2008). Is that paper really due today? Differences in first-generation and traditional college students’ understandings of faculty expectations. Higher Education, 55, 425–446.

23.

Cotton

Wilson

(2006). Student-faculty interactions: Dynamics and determinants. Journal of Higher Education, 51, 487–519.

24.

Daniel

Lobdell

Springate

Rayome

Bottoni

Doerr

Allen

(2009). Varying the frequency of intentional communication between student affairs personnel, first-year students, and their parents. NASPA Journal, 46, 412–430.

25.

Deil-Amen

(2011). Socio-academic integrative moments: Rethinking academic and social integration among two-year college students in career-related programs. Journal of Higher Education, 82, 54–91.

26.

Deutschlander

(2017). Academic undermatch: How general and specific cultural capital structure inequality. Sociological Forum, 32, 162–185.

27.

Dumais

(2006). Early childhood cultural capital, parental habitus, and teachers’ perceptions. Poetics, 34, 83–107.

28.

Downey

D. B.

(1995) When bigger is not better: Family size, parental resources, and children’s educational performance. American Sociological Review, 60, 746–761.

29.

Fry

(2011). Hispanic college enrollment spikes, narrowing gaps with other groups. Washington, DC: Pew Hispanic Center.

30.

Fry

Taylor

(2013). High school drop-out rate at record low: Hispanic high school graduates pass Whites in rate of college enrollment. Washington, DC: Pew Research Center.

31.

Gardner

de Bruijn

G. J.

Lally

(2011). A systematic review and meta-analysis of applications of the self-report habit index to nutrition and physical activity behaviours. Annals of Behavioral Medicine, 42, 174–187.

32.

Gemmill

Peterson

(2006). Technology use among college students: Implications for student affairs professional. Journal of Student Affairs Research and Practice, 43, 280–300.

33.

Greenman

Bodovski

Reed

(2011). Neighborhood characteristics, parental practices and children’s math achievement in elementary school. Social Science Research, 40, 1434–1444.

34.

Grodsky

Jackson

(2009). Social stratification in higher education. Teachers College Record, 111, 2347–2384.

35.

Grodsky

Jones

(2007). Real and imagined barriers to college: Perceptions of cost. Social Science Research, 36, 745–766.

36.

Guiffrida

(2006). Toward a cultural advancement of Tinto’s theory. The Review of Higher Education, 29, 451–472.

37.

Hamilton

(2016). Parenting to a degree: How family matters for college and beyond. Chicago, IL: University of Chicago Press.

38.

Hofer

B. K.

(2011). Student-parent communication in the college years: Can students grow up on an electronic tether? The Bulletin of the Association of College Unions International, 79, 36–41.

39.

Jack

(2016). (No) Harm in asking: Class, acquired cultural capital, and academic engagement at an elite university. Sociology of Education, 89(1), 1–19.

40.

Jaeger

(2009). Equal access but unequal outcomes: Cultural capital and educational choice in a meritocratic society. Social Forces, 87, 943–971.

41.

Jaeger

(2011). Does cultural capital really affect academic achievement? New evidence from combined sibling and panel data. Sociology of Education, 84, 281–298.

42.

Joas

(1996). The creativity of action. Chicago, IL: University of Chicago Press.

43.

Kim

Sax

(2009). Student-faculty interaction in research universities: Differences by student gender, race, social class, and first-generation status. Research in Higher Education, 50, 437–459.

44.

Kraft

Dougherty

(2013). The effect of teacher-family communication on student engagement: Evidence from a randomized field experiment. Journal of Research on Educational Effectiveness, 6, 199–222.

45.

Kraft

Rogers

(2014). Teacher-parent communication. Experimental evidence from a low-cost communication policy. Paper presented at the Society for Research on Education Effectiveness Spring Conference, Washington, DC.

46.

Lareau

(2011). Unequal childhoods: Class, race, and family life (2nd ed.). Berkeley: University of California Press.

47.

Lareau

(2015). Cultural knowledge and social inequality. American Sociological Association, 80(1), 1–27.

48.

Lareau

Weininger

(2008). Concerted cultivation continues: Class, culture, and child rearing In Lareau

Conley

(Eds.), Social class: How does it work? (pp. 118–151). New York, NY: Russell Sage Foundation.

49.

Lee

Bowen

N. K.

(2006). Parental involvement, cultural capital, and the achievement gap among elementary school children. American Educational Research Journal, 43, 193–218.

50.

Liu

(2009). Ordinal regression analysis: Fitting the proportional odds model using Stata, SAS and SPSS. Journal of Modern Applied Statistical Methods, 8, 632–645.

51.

MacDonald

Malatest

Assels

Baroud

Gong

(2009). Final impacts report foundations for success project (A report prepared by R. A. Malatest & Associates LTD). The Canada Millennium Scholarship Foundation. Retrieved from https://malatest.com/CMSF%20FFS%20-%20FINAL%20Impacts%20Report.pdf

52.

Melendez

(2010). The influence of parental attachment on the college adjustment of White, Black, and Latina/Hispanic women: A cross-cultural investigation. Journal of College Student Development, 51, 419–435.

53.

Miller

Prentice

(2016). Changing norms to change behavior. Annual Review of Psychology, 67, 339–361.

54.

Mitchell

Hughes

(2014). Demographic and instructor-student interaction factors associated with community college students’ intent to persist. Journal of Research in Education, 24, 63–78.

55.

Myers

(2012). Are discussions about college between parents and their high school children a college-planning activity? Making the case and testing the predictors. American Journal of Education, 118, 281–308.

56.

National Center for Education Statistics. (2010). Web Tables-Profile of Undergraduate Students: 2007-2008 (Table 3.11). Washington, DC: Author.

57.

National Center for Education Statistics. (2011). NCES Common core data state dropout and completion data file, school year 2008-2009 (Version 1a). Washington, DC: Author.

58.

National Survey of Student Engagement. (2007). Experiences that matter: Enhancing student learning and success. High-impact practices boost learning, involved parents no problem (Annual report). Retrieved from http://nsse.indiana.edu/NSSE_2007_Annual_Report/

59.

Pascarella

Terenzini

Hibel

(1978). Student-faculty interactional settings and their relationship to predicted academic performance. The Journal of Higher Education, 49, 450–463.

60.

Potter

Roksa

(2013). Accumulating advantages over time: Family experiences and social class inequality in academic achievement. Social Science Research, 42, 1018–1032.

61.

Rios-Aguilar

Deil-Amen

(2012). Beyond getting in and fitting in: An examination of social networks and professionally relevant social capital. Journal of Hispanic Higher Education, 11, 179–196.

62.

Roksa

Deutschlander

(2018). Applying to college: The role of family resources in academic undermatch. Teachers College Record, 120(6), 1–30.

63.

Roksa

Deutschlander

Whitley

(2016, November). The importance of considering parents and families for student transitions and belonging. ASHE Annual Meeting, Columbus, OH.

64.

Schwebel

Walburn

Klyce

Jerrolds

(2012). Efficacy of advising outreach on student retention, academic progress and achievement, and frequency of advising contacts: A longitudinal randomized trial. NACADA Journal, 32, 36–43.

65.

Scrivener

Sommo

Collado

(2009, April). Getting back on track: Effects of a community college program for probationary students (MDRC Report). New York, NY: Manpower Demonstration Research Corporation.

66.

Scrivener

Weiss

M. J.

(2009). More guidance, better results? Three-year effects of an enhanced student services program at two community colleges. New York, NY: Manpower Demonstration Research Corporation.

67.

Sewell

W. H.

Jr. (1992). A theory of structure: Duality, agency and transformation. American Journal of Sociology, 98(1), 1–29.

68.

Sheeran

Maki

Montanaro

Avishai-Yitshak

Bryan

Klein

. . . Rothman

(2016). The impact of changing attitudes, norms, and self-efficacy on health-related intentions and behavior: A meta-analysis. Health Psychology, 35, 1178–1188.

69.

Stephens

Hamedani

Destin

(2014). Closing the social-class achievement gap: A Difference-education intervention improves first-generation students academic performance and all students’ college transition. Psychological Science, 25, 943–953.

70.

Strauss

Volkwein

(2004). Predictors of student commitment at two-year and four-year institutions. The Journal of Higher Education, 75, 203–227.

71.

Stuber

(2011). Inside the college gates: How class and culture matter in higher education. Lanham, MD: Lexington Books.

72.

Swidler

(1986). Culture in action: Symbols and strategies. American Sociological Review, 51, 273–286.

73.

Tinto

(1975). Dropout from higher education: A theoretical synthesis of recent research. Review of Educational Research, 45, 89–125.

74.

Tinto

(1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). Chicago, IL: University of Chicago Press.

75.

Umbach

Wawrzynski

(2005). Faculty do matter: The role of college faculty in student learning and engagement. Research in Higher Education, 46, 153–184.

76.

U.S. Department of Commerce, Census Bureau, Current Population Survey. (2017). Digest of Education Statistics 2017 (Table 302.10, October Supplement, 2000 through 2016). Retrieved from https://nces.ed.gov/programs/digest/d18/tables/dt18_302.10.asp

77.

Valant

Loeb

(2015). Information, choice, and decision-making: Field experiments with adult and student school choosers (Working paper). Stanford, CA: Stanford University.

78.

Webber

Krylow

Zhang

(2013). Does involvement really matter? Indicators of college student success and satisfaction. Journal of College Student Development, 54, 591–611.

79.

What Works Clearinghouse. (2017). Procedures handbook (Version 4.0). Retrieved from https://ies.ed.gov/ncee/wwc/Handbooks

80.

Williams

Jones

Tukey

(1999). Controlling error in multiple comparisons, with examples from state-to-state differences in educational achievement. Journal of Educational and Behavioral Statistics, 24, 42–69.

81.

Wolf

Sax

Harper

(2009). Parental engagement and contact in the academic lives of college students. NASPA Journal, 46, 325–358.

82.

Yee

(2016). The unwritten rules of engagement: Social class differences in undergraduate academic strategies. The Journal of Higher Education, 87, 831–858.

83.

York

Loeb

Doss

(2018). One step at a time: The effects of an early literacy text messaging program for parents of preschoolers. Journal of Human Resources, 53, 0517-8756r.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB