Stage 1: Environment Scan

The first stage of the evaluation involved an environment scan. The purpose was to establish fundamental parameters of the scope of the research and inform the development of the second and third stages. It focused on establishing a robust snapshot of women and men's engagement in paid work before and after the birth of a baby.

A random sample of birth records was taken from the Register of Births held by the Department of Internal Affairs. This sample covered a six-month period from 1 December 2004 to 31 May 2005. Births to mothers aged under 18 years of age and recorded still births were removed from the sample. Name and address details of either the mother or father from these selected records were then entered into a database. A tele-matching process was used with assistance from Telecom. The overall matching rate was one in three.

Approximately 1750 records were matched and each person was sent a letter describing the research and providing an opportunity to decline to participate in the research. Twelve respondents (0.7 percent) withdrew from the research.

A total of 1000 interviews were conducted between 1 September 2005 and 2 October 2005. All interviews were conducted by telephone from Research New Zealand's CATI (Computer Assisted Telephone Interviewing) Unit in Wellington. The interviews were split between 720 mothers and 280 fathers, and both groups were asked the same set of questions (except in the case of single mothers). The overall response rate was 68 percent.

Stage 2: Qualitative Research

The second stage was qualitative research focused on uncovering the drivers of decision-making in relation to parental leave. The intention was to identify issues and insights that could be retested in Stage 3 to understand the size of the issues.

A total of 55 individual in-depth interviews, of approximately one and half hours were conducted in Wellington between 7 February and 10 March 2006. The interviews were conducted by Research New Zealand's Qualitative Unit. The sample consisted of:

• 24 mothers who were eligible for parental leave under the Act
• 11 spouses/partners of eligible mothers
• 8 mothers who were ineligible
• 12 employers.

All of the participants were from the greater Wellington region. The mothers interviewed spanned fi rst-time mothers and mothers who already had children (with a slight skewtowards new mothers), tertiary qualifi ed mothers and non-tertiary qualifi ed mothers. All mothers interviewed had a spouse/partner (i.e. no single mothers). As indicated in Stage 1, the profi le and paid work patterns of selfemployed tended to be similar to those of eligible mothers who took PPL. Therefore, it was decided to exclude this group. The employers interviewed spanned small, medium and large organisations and both public and private sector employers.

During Stage 2, the researchers identifi ed that there was some confusion about paid parental leave schemes offered by employers and the government-funded PPL under the Act. This highlighted the need to retest earlier fi ndings from the environment scan relating to the overall uptake of government-funded PPL. This area was retested in Stage 3.

Stage 3: Quantitative Survey

The third and fi nal stage of the research was a quantitative survey to test and quantify the key issues raised in the environment scan and qualitative stage of the research.

For this stage a random sample of 6,000 birth records was taken from the Register of Births covering babies born between March and May 2005 excluding participants in the environment scan, still births, and babies born to mothers aged under 18 years of age. Since Stage 1 showed no signifi cant variations for ethnicity, ethnic quotas were not set. Therefore sample sizes for Maori, Pacifi c and Asian mothers are small and it is not possible to note any areas of significant difference.

A tele-matching process was used with assistance from Telecom. The overall matching rate was one in three.

Approximately 2,200 records were matched. A total of 1501 mothers and 697 fathers were invited by letter to participate in the survey. These sample sizes were based on the number of interviews required for each group and the projected response rate. Each person was sent a letter describing the research and providing an opportunity to decline to participate in the research. From this group 151 people were ineligible [because of either not being in paid work during the six months prior to having a baby (124) not giving birth during the sample period (7) or declined to participate (30)]. All interviews were conducted by telephone from Research New Zealand's CATI (Computer Assisted Telephone Interviewing) Unit in Wellington. A separate questionnaire was used for each group. A total of 501 interviews were completed with mothers (response rate of 50 percent) and 150 interviews were completed with fathers (response rate of 29 percent).^[3] While some of the questions asked of mother and fathers were the same, or similar, some were tailored specifi cally towards mothers or fathers. While a household income variable was collected, the data on mothers and fathers were collected based on individuals not couples. Therefore, unlike some overseas research (e.g. Smeaton, 2006), the attitudes and behaviours of mothers and fathers cannot be directly linked to each other.

All interviews were conducted between 25 July and 6 August 2006. This meant that at the time of the interview parents had babies aged between 14 and 17 months.

With the introduction of PPL for the self-employed in July 2006, some self-employed people appear in the results of Stage 3 as eligible for PPL. While these respondents may not have been eligible for PPL at the time of the birth of their babies (between March and May 2005), when they were interviewed between 25 July and 6 August 2006 and had returned to work, some had become eligible for PPL due to the passing of amendments to the legislation extending PPL to the self-employed.

A sample of 2000 employers was drawn from a database held by Inland Revenue of organisations who have had at least one employee who has taken PPL during the last two years. Each organisation was sent a letter describing the research and providing an opportunity not to participate in the research.

Of the 22,000 organisations in Inland Revenue's database, 81 percent had only one employee take government-funded PPL, 1 percent had between two and fi ve employees and 3 percent had more than fi ve employees. To ensure that organisations with a large number of employees taking parental leave were included in the research, this group was over-sampled. A total of 150 interviews were completed with employers (response rate of 17 percent).

Quotas were established to provide that an equal number of small (under fi ve employees), medium (fi ve to 19 employees) and large (over 20 employees) organisations were interviewed. The results were subsequently re-weighted back into their true proportions in New Zealand (i.e. small 63 percent, medium 27 percent and large 10 percent). One-third of employers were male, and the other two thirds were female. One-third of employers (35 percent) had female employees only. These were more likely to be organisations of less than fi ve employees. A further 18 percent had between 90-99 percent female employees.