Teen Parties: Who Has Parties, What Predicts Whether There is Alcohol and Who Supplies the Alcohol?

Abstract

This study explores which youth are more likely to have parties at home, what factors are associated with the presence of alcohol at parties, and who supplies the alcohol. We collected data in 2011 and 2012 through telephone interviews with 1,121 teens living in 50 mid-sized California cities. Overall, about a quarter of teens reported having had a party at their house in the past 12 months, of whom 39% reported that there was alcohol at their last party. Multiple sources supplied alcohol for most parties. Seventy-two percent of those having a party stated that at least one of their parents knew about their last party, and 64% reported that a parent was home at least part of the time. Seventy percent of youth who hosted a party with alcohol said that their parent(s) definitely knew that there was alcohol at the party, 24% replied that their parent(s) probably knew, and only 5% said that their parent(s) did not know that there was alcohol at the party. Logistic regression analyses indicated that youth with parents who host parties at home are themselves more likely to host parties at home. Having alcohol at a party was positively related to the age of the teen and the number of guests attending, and was negatively related to parents' awareness of the party. However, we found no relationship between whether a parent was at home at the time of the party and whether it included alcohol. These findings suggest that teens who have parties with alcohol at home have parents who know that there is alcohol at the party, even though only a small number of parents provided alcohol for the party.

Introduction

Youth drinking is an important public health issue. The Monitoring the Future (MTF) survey, a national study that includes approximately 50,000 8^th, 10^th and 12^th graders each year, indicates that 62% of all high school seniors report drinking in the past year and 39% report doing so in the past 30 days (Johnston, O'Malley, Bachman, & Schulenberg, 2013). A quarter of high school seniors report being drunk at least once in the past month.

Parties, often hosted in homes, are a common location where teens engage in drinking (Paschall, Grube, Black, & Ringwalt, 2007). Teen parties are a high risk setting for youth (Mayer, Forster, Murray, & Wagenaar, 1998; Schwartz & Little, 1997) and heaviest drinking occurs when teens are with peers and in large groups with other youth their age (Mayer, Forster, Murray, & Wagenaar, 1998). Furthermore, involvement in social activities with peers, such as parties, is associated with a youth being more likely to move from normative to high risk drinking (Power, Stewart, Hughes, & Arbona, 2005).

Parties that involve teen drinking may occur when parents are not at home. However, drinking at parties in the home may be sanctioned or ignored by some parents who believe it is safer for teens to drink in a supervised setting (Dills, 2010). The purpose of social host ordinances is to deter parties in private residences that include underage drinking. These ordinances can impose criminal and/or civil sanctions for adults who provide alcohol to minors on property they own, lease, or otherwise control. In California, where we conducted this study, there is no state-wide social hosting law. However, the Teen Alcohol Safety Act of 2010 allows civil lawsuits against an adult host who provides alcohol to teens. In addition, many municipalities have established local social host ordinances. About half of the cities (24 out of 50) from which the youth were sampled for this study have such ordinances (Thomas, Paschall, Grube, Cannon, & Treffers, 2012).

The evidence that social host ordinances reduce alcohol use by minors or problems related to underage drinking is mixed. A recent study found that youth are less likely to drink in large peer groups in communities with social host ordinances, but that these ordinances had no effect on heavy episodic drinking (Wagoner et al., 2013). Another study found that social host ordinances may lower the drunk-driving fatality rate among 18-20 year olds (Dills, 2010). However, still another study found no relationship between social host laws and fatal car crashes (Fell, Fisher, Voas, Blackman, & Tippetts, 2007). The ineffectiveness of social host and other underage drinking ordinances has been attributed to municipalities that do not follow best practices or lack effective enforcement (Thomas et al., 2012). Lack of awareness on the part of parents of social host ordinances may also be a contributing factor.

Even though parties are a common drinking context for adolescents, little is known about those who have parties at home and what characteristics are related to whether there is alcohol at the party. In this study we examined which youth are more likely to have parties, what predicts whether there will be alcohol at the party, and who supplies the alcohol.

Method

Survey data

Survey procedures

We collected data in 2011 and 2012 through telephone interviews conducted for the third wave of an annual longitudinal survey study of youth (n = 1,121) living in 50 mid-sized (populations between 50,000 and 500,000) California communities (see Paschall et al., 2012 for more details on survey methodology). Not included were urban areas such as Los Angeles, San Francisco, and San Diego, small towns with populations under 50,000, or rural areas. We focused on mid-sized cities because most cities in the U.S. are within this population range. We excluded larger urban areas because they tend to be heterogeneous in terms of population and may have unique land use characteristics, such as ports. We excluded smaller communities because the rates for some problem outcomes that were the focus of the main study (e.g., single vehicle nighttime crashes) are very low.

Initially, we identified households through a purchased list-assisted sample of addresses and phone numbers for the 50 cities. All selected households received a letter notifying them that they would be contacted by telephone and invited to participate in a telephone survey if they met the selection criteria. During the telephone call, interviewers screened households for teens between the ages of 13 and 16, the target age for Wave 1. If there was more than one youth in the target age range in the home, the youth with the most recent birthday was selected. Interviewers obtained parental consent for the interviews followed by assent from the youth respondents. The telephone interviews took approximately 30 minutes.

Interviewers contacted a total of 15,694 households to obtain a sample of 3,062 households with eligible respondents. Of the 3,062 households, 1,543 (50%) participated in Wave 1. At Wave 3, a total of 1,121 of these respondents participated (73% retention rate). Of the 1,121 respondents, 1,055 provided complete data for all variables included in the analyses for this paper, and 272 reported having had a party at their home. Of these 272, 106 reported that there was alcohol at the party. Youth who reported having had a party at home and those who reported that there was alcohol at the party constitute the samples for this study.

Background variables

The survey asked respondents about their race/ethnicity. We recoded their responses into a dichotomous variable (0 = Non-White, 1 = Non-Hispanic White). The survey asked gender (0 = female, 1 = male), and age (a continuous variable). The survey also asked youth about their mother's and father's highest level of schooling completed (0 = neither parent had a college degree or higher, 1 = at least one parent had a college degree or higher).

Teens hosting parties at home

The survey asked respondents how many parties they had at their home in the past 12 months. Interviewers instructed youth not to count parties given by their parents/guardians or other relatives or family events. We recoded this variable into a dichotomous variable (0 = no, 1 = yes). Interviewers then asked youth how many people attended the last party they had at home.

Alcohol at the party

Those youth who reported having had a party were asked whether there was alcohol at the last party at their home (0 = no, 1=yes). Those who responded affirmatively were asked who provided the alcohol: self, their parent(s), guests, or someone else. Respondents could indicate multiple sources.

Parental supervision

The survey asked youth whether at least one parent knew that they were having a party (0 = no, 1 = yes), and whether at least one parent was at home at the time of the party (1 = no parents at home at any time during party, 2 = at least one parent at home part of the time, 3 = at least one parent at home for the entire time). We recoded the responses into a dichotomous variable (0 = no parent(s) at home, 1 = at least one parent at home for part or all of the time). The survey asked youth who reported that there was alcohol at the party whether their parent(s) knew that there was alcohol, and we coded responses into a dichotomous variable (0 = neither parent knew there was alcohol, 1 = at least one parent probably or definitely knew). When youth responded that their parents did not know that they had a party, interviewers did not ask this question.

Other parties attended

The survey asked youth how many parties they went to in the past 12 months (a continuous variable), and at how many of the parties they attended kids their age were drinking alcohol (0 = none of them to 4 = all of them).

Parents hosting parties at home

We asked respondents how often their parents had a celebration or party at home in the past 12 months (0 = never, 1 = a few times but less than once a month, 2 = about once a month, 3 = a few times a month, 4 = about once a week, 5 = two or more times a week).

Social host ordinances

An overall social host ordinance score was calculated for each community based on the presence of specific recommended elements (Thomas, Paschall, Grube, Cannon, & Treffers, 2012): social host law (+1); applies to underage parties (+2); civil liability (+2); criminal liability (+1); ordinance applies to a range of property types (residence, outdoor property) (+1); and knowledge requirement (parents have to be aware that a party is taking place on their property) (-1). Scores ranged from 0 to 7. Data on these policies were gathered from cities' website and City Clerks (see Thomas et al., 2012).

Analyses

We conducted logistic regression analyses to examine the relationships between individual level factors, whether the teen had a party and whether alcohol was present at the party. Although respondents were nested in cities, due to the small number of youth having parties we were unable to analyze the data using hierarchical linear modeling. Although the ICCs were small (.000 and .005 for having parties and for having alcohol at the party, respectively), we weighted the data for design effects to account for nesting within communities (Kish, 1965) and included the social host ordinance score as an individual-level variable.

Results

Sample characteristics

We display the characteristics of the study sample in Table 1.

Table 1. Demographics of respondents (n=1,121).

	Youth Survey
Race
White	71%
African American	3%
Asian	5%
Hispanic	24%
Filipino	2%
Native American	.6%
Native Hawaiian/Pacific Islander	.4%
Other/multi-racial	19%
% of parent(s) who completed college or higher	70%
Youth Gender
Female	46%
Youth Age (mean)	16.8
Age 15	12%
Age 16	28%
Age 17	31%
Age 18	23%
Age 19	6%

Teens hosting parties at home

Twenty-four percent of teens responded that they had a party at their house in the past 12 months. These teens reported a mean of about three parties (M = 2.9, SD = 4.5). The last party they hosted was attended, on average, by about 20 people (M = 20.1, SD = 29.9).

Alcohol at teen parties

More than one third (39%) of youth who hosted parties reported that there was alcohol at their last party. Sixteen percent of parties hosted by 15 year olds had alcohol, a percentage that increased to 22% for 16 year olds, 38% for 17 year olds, 61% for 18 year olds, and 82% for 19 year olds. Most of those who had alcohol at their last party reported multiple sources of alcohol. About a quarter (23%) of respondents who hosted parties with alcohol reported only one source of alcohol for their last party, half (51%) reported two sources, and a quarter (26%) reported three or more sources. Of those who reported having alcohol at their last party, 38% reported that they provided alcohol themselves, 84% replied that guests brought the alcohol with them, 74% responded that someone else provided it, and 9% replied that their parents provided it.

Parental supervision

Seventy-two percent of respondents who had a party replied that at least one of their parents knew about their last party, and 64% of them reported that at least one parent was home at least part of the time. Seventy percent of youth who reported having alcohol at their party indicated that at least one of their parents definitely knew that there was alcohol at the party, 24% replied that at least one of their parents probably knew, and 5% said that their parents did not know. Neither parents' awareness of the party (r = -.03) nor their presence at the party (r = -.05) was significantly related to the number of guests who attended.

Other parties

We asked all youth, regardless of whether they had hosted a party, about their party attendance in the past year. Three-quarters (76%) of the youth reported going to at least one party, and attended an average of 8.7 parties (SD = 13.1). About one-third (31%) of youth who had attended at least one party reported that teens drank alcohol at all of the parties, 16% reported that teens drank at most of them, 9% said teens drank at some of them, 16% said a few of them, and 28% reported that youth their age did not drink at any of the parties they had attended.

Parents hosting parties at home

Seventy-six percent of teens responded that their parent(s) had hosted at least one party or celebration at their house in the past 12 months. Fifty-seven percent of respondents reported that parents hosted parties a few times but less than once a month, 12% reported about once a month, and 7% reported that their parents had parties or celebrations a few times a month or more often.

Social host ordinance score

The social host ordinance scores for communities in which respondents resided ranged from 0 to 7 (M = 2.7, SD = 3.1). Just over half (54%) of the respondents lived in a city without a social host ordinance.

Multivariate analyses

Who hosts parties at home?

Our first model in which we included age, gender, race, parents' education, parents hosting parties at home, and social host ordinance score, examined characteristics associated with hosting a party at home (Table 2). We found a positive relationship between the number of parties hosted by parents and youth who hosted their own parties at home.

Table 2. Summary of logistic regressions predicting party at home and alcohol at party.

	b	SEb	Wald	p	OR
Hosting a party at home
Male	-.05	.15	.11	.739	.95
Age	.06	.07	.59	.443	1.06
White	.08	.16	.24	.623	1.08
Parents have college degree	.32	.18	3.21	.073	1.37
Parents hosting parties	.35	.09	15.33	.000	1.41
Social host ordinance	.01	.03	.175	.676	1.01
Alcohol at the party*
Male	-.27	.42	.43	.513	.76
Age	.59	.20	8.74	.003	1.81
White	-.78	.45	3.05	.081	.46
Parents have college degree	-.48	.46	1.08	.299	.62
Social host ordinance	.004	.07	.003	.957	1.00
Number of guests	.06	.02	9.36	.002	1.06
Parents knew about party	-2.94	.64	21.26	.000	.05
Parents at home	-.99	.55	3.28	.070	.37

^*limited to youth who have hosted a party at home

What predicts alcohol at a party?

We included age, gender, race, parents' education, social host ordinance score, parent being at home, parent knowing about party, and number of guests in the model (Table 2). We found a positive relationship between the teen's age, the number of guests and the presence of alcohol. If parents knew about the party it was less likely to have alcohol, but there was no relationships between whether a parent was at home at the time of the party and the presence of alcohol.

Discussion

About one quarter of the adolescents in our survey reported that they had a party at their home in the past year. Only one individual-level characteristic predicted which teens hosted parties at home, namely the frequency with which parents hosted parties and celebrations at home. We suspect that social parents may be more likely to have social teens; it may also be possible that parents who host parties at home are more permissive of their teens doing the same.

More than a third of teens (39%) who hosted a party at home reported that there was alcohol at their last party. Older teens were more likely to have alcohol at their parties. Furthermore, when there was alcohol at a party, almost three quarters of parents knew about it. However, alcohol was less likely to be present at a party when at least one parent knew about the party. Surprisingly, there was no relationship between whether a parent was at home at the time of the party and whether alcohol was present there. There was a positive relationship between the number of guests and having alcohol present. Considering that guests often bring alcohol to the party, it is not surprising that having more guests increases the chances of alcohol being present. Having multiple sources of alcohol also suggests that it may be difficult for parents, even when they are in the home, to keep a party alcohol-free.

We found no relationship between the social host ordinance score and the likelihood that teens would host a party at home or that it would include alcohol. Social host ordinances may not be sufficiently enforced to have an impact on parents' or youth's behavior. Alternatively, parents may be unaware of social host ordinances and the possible legal consequences of allowing minors to drink in their home.

This study has several limitations. The number of respondents is fairly small because only about a quarter of the overall sample reported hosting parties at home. Because this study's sample is limited to youth who live in mid-sized cities in California, our findings may not be generalizable to large metropolitan or rural areas. Furthermore, some groups are overrepresented in this sample compared to the general population in California; in particular, our sample comprised fewer minorities and was more highly educated. We recognize that other individual-level characteristics that we did not consider - such as whether respondents' parents rented or owned their home – may play a role in whether a teen hosts a party at home. An additional limitation is that we were unable to address the amount of alcohol consumed at the parties. Nor do we know the proportion of guests involved in drinking and supplying alcohol.

Nonetheless, this study provides new information about parties hosted by teens. In particular, we found that parents may have some influence over their teens' parties. The children of parents who knew about the party were less likely to serve alcohol. However, we found no relationship between parents being at home at the time of the party and alcohol being served. This finding suggests that parents who permit (or are aware of) a party may be more likely to condone the presence of alcohol. We also found a greater likelihood of having alcohol at the party when there were more guests. However, we found no relationship between parents either knowing about or being present at the time of the party and the number of guests in the home, suggesting that parental monitoring is not an effective strategy to manage teen parties.

Our study highlights the need to better inform parents about their liabilities for allowing minors to drink in their homes. In addition, we need more in-depth understanding about how parents can keep teen parties alcohol free. We also need to learn more about parents who allow alcohol at parties; for example, their motivations for doing so. Furthermore, we need to examine what, if anything, parents know about social host ordinances and whether their understanding of these ordinances and related penalties might deter parents from allowing their teens to have parties at home at which alcohol is present.