View in EDS HTML Full Text PDF Full Text

Life Course Productivity Model to Analyze Academic Research Issues: A Longitudinal Analysis at One Taiwanese University

Saved in:
Bibliographic Details
Title: Life Course Productivity Model to Analyze Academic Research Issues: A Longitudinal Analysis at One Taiwanese University
Language: English
Authors: Fu, Yuan-Chih (ORCID 0000-0003-1397-2207), Chan, Sheng-Ju (ORCID 0000-0003-3830-5304), Huang, Shi-Ming (ORCID 0000-0001-6062-7153), Lee, Ya-Hui (ORCID 0000-0001-5802-9146)
Source: Studies in Higher Education. 2021 46(11):2491-2505.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 15
Publication Date: 2021
Document Type: Journal Articles
Reports - Research
Education Level: Higher Education
Postsecondary Education
Descriptors: Foreign Countries, Research Universities, College Faculty, Teacher Researchers, Productivity, Institutional Research, Reinforcement, Noninstructional Responsibility, Age
Geographic Terms: Taiwan
DOI: 10.1080/03075079.2020.1723535
ISSN: 0307-5079
Abstract: Research productivity has been a critical issue in terms of academic development in higher education. In this study, we adopt a life-course perspective to examine the personal factors, mostly age-related, affecting research productivity in a Taiwanese research-oriented university. Covering a time series of 20 years, our dataset includes individual research performance of faculty and other relevant covariates over their life course. The growth curve model designed for multilevel modeling of repeated measures is applied to capture the age effect. Our analysis contributes to the thread of this literature in several dimensions. First, the faculty's early academic achievement is positively associated with their later performance providing support for the cumulative advantage theory. Unlike the prediction of the utility maximizing theory, faculty with an administrative position leads to higher productivity. Finally, reinforcement still plays a critical role in regulating the productivity for non-early promising faculty.
Abstractor: As Provided
Entry Date: 2021
Accession Number: EJ1316346
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGMG5QjKB0Ot-ybAlju-sQwAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDJ5AedOKEoB4_LuOOQIBEICBm5301adtRBBS1sbC3O1RT4D0i1y-BALoPXrlKra32MxGspg_TGQgelVmD5TkokV9pp2VFSoA6ssFCDYzIxvPsHcM0ixeEgwA0aKXZjAQGYyL1SefmvVi1ES7BDVUqt6GovgGWpGm-YLEihNQShjFJEkX24Isy1yCvVCth2pPOuNC1hwE9SEIKy_uu8DeTRpnilHVaW8poVSeek3W
Text:
  Availability: 1
  Value: <anid>AN0153185467;she01nov.21;2021Oct26.06:46;v2.2.500</anid> <title id="AN0153185467-1">Life course productivity model to analyze academic research issues: a longitudinal analysis at one Taiwanese university </title> <p>Research productivity has been a critical issue in terms of academic development in higher education. In this study, we adopt a life-course perspective to examine the personal factors, mostly age-related, affecting research productivity in a Taiwanese research-oriented university. Covering a time series of 20 years, our dataset includes individual research performance of faculty and other relevant covariates over their life course. The growth curve model designed for multilevel modeling of repeated measures is applied to capture the age effect. Our analysis contributes to the thread of this literature in several dimensions. First, the faculty's early academic achievement is positively associated with their later performance providing support for the cumulative advantage theory. Unlike the prediction of the utility maximizing theory, faculty with an administrative position leads to higher productivity. Finally, reinforcement still plays a critical role in regulating the productivity for non-early promising faculty.</p> <p>Keywords: Scientific productivity; life course; growth curve model; graying academia; institutional research</p> <hd id="AN0153185467-2">Introduction</hd> <p>For the past decades, Taiwan's higher education sector has seen a rapid expansion in terms of size and scale. Massification and even universalization have become major outcomes of such grand transformation (Chan [<reflink idref="bib4" id="ref1">4</reflink>]). Concomitantly, postgraduate programs are spreading across university campuses. Faculty are gradually engaging in research activities and knowledge production. If the growth of research output and publication is examined, there has been an accelerating trend since the 1990s (Scimago Institutions Rankings [<reflink idref="bib19" id="ref2">19</reflink>]). Such progress is mainly propelled by two major factors: the enlarged amount of research faculty and the enhancement of research productivity. The former is related to the massification while the latter is associated with the maturity of the knowledge production sector. However, the faculty within Taiwan's higher education institutions are aging because of low retirement rates and a decline in the enrollment population of the college cohort. The low retirement rate today is the result of the hiring bulges of the 1990s as well as a recent reform on the pension policy. According to the official statistics from the Ministry of Education in Taiwan, 31.1% of the Taiwanese faculty was more than 55 years old in 2017, while up to 54.4% was older than 50. This percentage is even higher at research-intensive universities indicating an unbalanced composition of faculty as a whole in terms of organizational health. Such faculty configuration has made policymakers become aware of the possible consequences of an aging campus. Despite this realization, the extremely low faculty replacement rate along with the very limited vacancy positions for young scholars still makes it hard to change the aged population of college faculty.</p> <p>The consequences of the graying Taiwanese academic community, particularly as it relates to the age-research productivity relationship, can be examined in several ways. First, research activity is an intellectual task requiring high-intensity investment of time and energy. If age plays a determining role on individual research productivity, policymakers at the national level should prepare incentives that ensure the stable replacement of faculty instead of deterring them from retiring. Second, given the importance of research in establishing institutional reputation and its role in the awarding of grants, if age is related to productivity then an aging academic community may directly impact the resources available to a university.</p> <p>The inquiry on the impact of aging on a faculty's research profile is not a new topic. A popular belief held by scientists and the lay public alike is that science is a young person's game (Levin and Stephan [<reflink idref="bib14" id="ref3">14</reflink>]). Since the 1970s, a thread of empirical studies almost concluded that age does play a role on research productivity (Bayer and Dutton [<reflink idref="bib2" id="ref4">2</reflink>]; Levin and Stephan [<reflink idref="bib14" id="ref5">14</reflink>]; Rauber and Ursprung [<reflink idref="bib17" id="ref6">17</reflink>]; Sabharwal [<reflink idref="bib18" id="ref7">18</reflink>]). A well-established theory of research productivity is the life-course model, which suggests that the output will become hump-shaped over the period of a researcher's academic career. However, some weaknesses exist in past empirical studies. First, most of the findings are based on cross-sectional data and did not track the life course of individuals. The survey data only allowed them to construct datasets based on different faculty at different career or age stage (Tien and Blackburn [<reflink idref="bib21" id="ref8">21</reflink>]; Hu and Gill [<reflink idref="bib10" id="ref9">10</reflink>]; Sabharwal [<reflink idref="bib18" id="ref10">18</reflink>]). Second, even though some of the studies did construct a longitudinal dataset, they did not always include relevant covariates that could have either positive or negative impacts on research productivity (Goodwin and Sauer [<reflink idref="bib8" id="ref11">8</reflink>]). Neglecting the existence of these covariates could bias the estimation of the aging effect. Third, some of the previous studies refer the faculty's research productivity as seen through academic publications (Levin and Stephan [<reflink idref="bib14" id="ref12">14</reflink>]; Perry et al. [<reflink idref="bib16" id="ref13">16</reflink>]). However, the amount of personal publications reported by survey participants is not always reliable, particularly in such long periods. The amount of granted research, which also passes through the peer-review process (Sabharwal [<reflink idref="bib18" id="ref14">18</reflink>]) and is recorded by the institutional data system, is a quality proxy to reflect the research productivity. It is fair to say that the aging effect should be re-examined by taking advantage of longitudinal administrative data, which could precisely capture the growth curve of faculty's research profiles. Additionally, we will employ life course theories (i.e. cumulative advantage, utility maximizing, and obsolescence theory) to examine how individual researchers develop their productivity.</p> <p>Furthermore, heterogeneity is always of interest in a study. Casual observation suggests that the temporal pattern of productivity still varies across individuals. Under the life-course pattern, faculty with different characteristics share different life trajectories. The widely investigated characteristics such as gender and research fields almost reached a consensus (Becher [<reflink idref="bib3" id="ref15">3</reflink>]; Stack [<reflink idref="bib20" id="ref16">20</reflink>]; Wanner, Lewis, and Gregorio [<reflink idref="bib23" id="ref17">23</reflink>]). However, while people tend to assume that foreign-educated scholars should be superior to their colleagues who were conferred with a domestic PhD degree, there is no evidence to prove or refute this stereotype. The last of the issues concerns young scholars who have a very promising start at the early stage of career. How different would their research profile be from their peers throughout their career path? Could they keep their momentum and be less affected by the aging effect? This study is situated in a Taiwanese research-oriented university that is facing a decline in its amount of research productivity and is therefore urged to find out the cause behind this scenario. The query to these questions based on the aging effect perspective not only contributes to this thread of literature but also has its policy implication on the sustainability of university management.</p> <hd id="AN0153185467-3">Literature review</hd> <p></p> <hd id="AN0153185467-4">Major theoretical framework: a life-course perspective</hd> <p>One of the well-established theories of age-research productivity relationship is the life-course model. Adopting a 'life-long' approach suggests that individuals may present various features and traits at different life stages in terms of research productivity. As we would like to investigate the age-related variables at the personal level, adopting the life-course perspective can help to understand the changing dynamics of research productivity in a longitudinal manner. The most prevailing and relevant theories include cumulative advantage, utility maximizing and the obsolescence theory. In fact, they focus on different phases of the life stage with various implications to research productivity. In other words, these various theories can examine the different influences of age-related variables at the conceptual level offering more insights into changing productivity.</p> <p>The cumulative advantage theory proposes that earlier academic success may lead to later productivity. This means that if a young scholar can have higher and better publications, he or she will have higher opportunities in producing more academic papers. This is because earlier positive outcomes tend to accumulate successful experiences, extra grants, social networks, and research facilities and equipment. These cumulative advantages can empirically sustain later productivity in academic life. This theory, therefore, emphasizes the causal relationship of research productivity between earlier and later career. Put differently, as said by well-known scholar, Robert K. Merton, in his seminal study on the Matthew effect, this includes 'the accruing of greater increments of recognition for particular scientific contributions to scientists of considerable repute and the withholding of such recognition from scientists who have not yet made their mark' (Merton [<reflink idref="bib15" id="ref18">15</reflink>], 58). This, then, is a broader concept of 'reinforcement' conditioning the upcoming performance of individual researchers. This means that a good start can potentially stimulate the success of the later life stage. Thus, this theory based on the notion of cumulative advantage and 'reinforcement' will be employed to examine their validity within the empirical situation.</p> <p>The utility maximizing theory denotes that individual scholars might maximize their utility by allocating their priority to different types of works including research, teaching or administration, or professional activities. Individual faculty would, therefore, choose their priority based on the maximal utility they can gain from a variety of jobs. As Kyvik ([<reflink idref="bib12" id="ref19">12</reflink>]) pointed out, 'both competent and less competent researchers will choose to reduce their research efforts over time, because they think other tasks may be more advantageous' (<reflink idref="bib40" id="ref20">40</reflink>). Therefore, faculty might spend more time in administrative jobs or professional activities outside the university once they think further effort in research cannot increase their chances of rewards. The utility maximizing theory points out two main ideas: along with the age, productivity will decrease and faculty will choose to engage in non-research activities for boosting long-term benefits (Kwiek [<reflink idref="bib11" id="ref21">11</reflink>]). This study, employing teaching and administrative variables within its framework, will examine whether such assumption is valid or not.</p> <p>The obsolescence theory directly points out that senior faculty suffer from declining research productivity. This setback is not due to the decreasing 'marginal utility' (i.e. more research doesn't yield the same rewards proportionally in the long run), as the utility maximizing theory indicated, but resulted from the general decline in intellectual abilities that come with aging (Sabharwal [<reflink idref="bib18" id="ref22">18</reflink>]). In other words, older faculty are not in line with the cutting-edge development of knowledge production within the context of major technological advances. This can be true for the life sciences, ICT and even physics. Levin and Stephan ([<reflink idref="bib13" id="ref23">13</reflink>]), using this theory, explained why those physicists trained at a time when S-Matrix theory was popular were displaced by the new quantum theory. They found themselves with an obsolete stock of intellectual capital due to the rapid transformation of the 'half-life of knowledge.' We will, therefore, estimate whether the obsolescence theory is empirically supported by our investigated cases.</p> <hd id="AN0153185467-5">Research productivity and various characteristics</hd> <p>Several issues regarding the aging effect have been well investigated. The first is when the researcher would reach their peak of productivity. Using physicists as research samples, Cole's ([<reflink idref="bib6" id="ref24">6</reflink>]) and Bayer and Dutton's ([<reflink idref="bib2" id="ref25">2</reflink>]) studies imply that output peaks at around age 40. Sabharwal ([<reflink idref="bib18" id="ref26">18</reflink>]) also found that the majority of the research output produced is by early and mid-career faculty members. However, Levin and Stephan's ([<reflink idref="bib14" id="ref27">14</reflink>]) study indicates that the output of an older group is always less than the youngest, which is inconsistent with the hump-shaped growth curve assumption.</p> <p>Although the life-course model has been accepted as the foundational base to understand the trajectory of research productivity, casual observation suggests that the temporal pattern of productivity still varies across individuals. Apparently, faculty with different characteristics have different life trajectories. Those differences have captured the attention of policymakers and researchers. Among those characteristics, the discipline, gender, career choice, educational background, and working environment are widely investigated. The impression about the differences of research productivity by discipline is that natural science would exceed social science. Wanner, Lewis, and Gregorio ([<reflink idref="bib23" id="ref28">23</reflink>]) confirmed this observation by using 1972–1973 national survey data while publications were the proxy for individual productivity. Specifically, the publication rates for natural scientists exceeded social scientists by about 60 percent. Becher ([<reflink idref="bib3" id="ref29">3</reflink>]) argued that the lower number of articles produced by social scientists is in part a reflection of the nature of the discipline: longer publication time, lengthier articles, fewer grants, and the difficulty of obtaining data. Gender is another factor affecting researchers' research performance (Holliday et al. [<reflink idref="bib9" id="ref30">9</reflink>]). Female faculty tend to be less productive compared to their male colleagues. Stack ([<reflink idref="bib20" id="ref31">20</reflink>]) pointed out that care-giving responsibilities, childbearing, and the age of children are all factors that take time away from research resulting in an aggregate disadvantage accumulated by female faculty over time further lowering their research productivity. This phenomenon could be more apparent in Asian countries where the society widely assumes that the mother should bear more family responsibilities even when taking on highly professional positions.</p> <p>Another thread of empirical studies focuses on the working environment where the faculty is affiliated. Goodwin and Sauer ([<reflink idref="bib8" id="ref32">8</reflink>]) investigated the publication histories of academic economists and found that those who took academic administrative positions, such as department head or dean, showed a significant drop in productivity compared to their colleagues. Similar to the teaching load which had an adverse effect on research productivity (Hu and Gill [<reflink idref="bib10" id="ref33">10</reflink>]), an administrative position is assumed to be a deterrent which would decrease the faculty's time invested on research. However, there are also some factors bolstering the faculty's performance. Faculty who teach in graduate programs are more productive than those who do not (Levin and Stephan [<reflink idref="bib14" id="ref34">14</reflink>]). This is because graduate students, doctoral students particularly, are expected to engage in research projects and provide valuable assistance to their advisors. Hu and Gill's ([<reflink idref="bib10" id="ref35">10</reflink>]) study showed that for those senior faculty members in the schools where IS doctoral programs are offered, their productivity is less affected by aging. Put differently, the declining productivity due to the aging effect of faculty is countered with the involvement of doctorial students.</p> <p>Since the life-course model is based on the incentive mechanism modulated by the process of natural aging, changing the schedule of reinforcement, such as the promotion system, has become the interest of higher education leaders. For instance, Tuckman ([<reflink idref="bib22" id="ref36">22</reflink>]) proposed that the academic reward structure should be changed by increasing the number of ranks that faculty can attain in response to the graying American campus. In many countries, this theory sets up the rationale for a faculty ranking system shifting from a seniority-based to a merit-based promotion system. In 1998, the Taiwan ministry of education changed the ranking system by adding a new rank level, thereby lengthening the faculty's career. Recently, the so-called distinguished professor title was also created to encourage the faculty who had already obtained professorship to keep their momentum in research. Using a Taiwanese example, Tien and Blackburn ([<reflink idref="bib21" id="ref37">21</reflink>]) did not find enough evidence to prove the presence of a reinforcement schedule.</p> <p>Several weaknesses exist in the previous empirical studies concerning the focus of this study. First, most findings are based on cross-sectional data, which did not track the life-course of individuals. The survey data only allowed them construct a dataset based on the different faculty at different career or age stages (Levin and Stephan [<reflink idref="bib14" id="ref38">14</reflink>]; Hu and Gill [<reflink idref="bib10" id="ref39">10</reflink>]; Sabharwal [<reflink idref="bib18" id="ref40">18</reflink>]). Second, even though some of the studies could construct a longitudinal dataset, they do not always include other relevant covariates which could have either positively or negatively impacted the research productivity (Goodwin and Sauer [<reflink idref="bib8" id="ref41">8</reflink>]). Neglecting the existence of these covariates could bias the estimation of the aging effect. It is fair to say that the aging effect should be re-examined by taking advantage of longitudinal administrative data, which could precisely capture the growth curve of faculty's research profiles. Our research can overcome the weaknesses of previous studies by using one university's faculty longitudinal data and including influential covariates such as academic field, rank, teaching load, supervision, and administrative position.</p> <hd id="AN0153185467-6">Methodology</hd> <p></p> <hd id="AN0153185467-7">Setting</hd> <p>This study is situated in a Taiwanese research-oriented university. It consists of seven colleges and has different academic fields including humanities, natural sciences, social sciences, engineering, management, law, and education. This university was established during the expansion of higher education in the late 1980s and since then it has seen many changes in terms of its faculty age and scale (Figure 1). In 1996, the number of faculty was 223. By 2006, the faculty number had almost doubled at 413 with 75% of them being between the ages of 35–49. Since then the scale of faculty was already saturated. By 2016, the number of faculty was 440, and those whose age was between 35–49 years old only accounted for 43% of whole full-time faculty. Over the past two decades, this university experienced a rapid expansion followed by the saturation of its faculty positions. Consequently, the low replacement rate of faculty has led to the graying of the faculty population, echoing the situation of the majority of higher education institutions in Taiwan as we have indicated before.</p> <p>PHOTO (COLOR): Figure 1. Pyramid of faculty population.</p> <hd id="AN0153185467-8">Data description</hd> <p>The data used in this study are the administrative data records approved and provided by the institutional research office at a research-oriented university covering the period from 1997–1998 academic year through 2016–2017. The administrative data warehouse collects and stores the institutional operation data over the past 20 years. Variables include full time faculty demographics and granted research project records. We refer to the personal research productivity as the amount of granted research projects received by faculty annually. Given the fact that only full-time faculty are qualified to apply for research grants from Ministry of Science and Technology (earlier named as the National Science Foundation, or NSF) and the review process would take more than half year, we limited our sample to full-time faculty members who were affiliated to this university for more than four academic years.</p> <p>Since we are interested on the impact of the aging effect on the personal research productivity, those full-time faculty members who were never granted with research projects were removed because their performances had nothing to do with their age. In order to observe the aging effect from their early to later career stage, we limited our faculty samples to those who were born during the 1960s. Thus, we construct a core sample comprising of 288 faculty members with 4039 observation points. Among 288 faculty, the maximum length of service is 20 years while the minimum is 4, with the average being 13. In 1997, the age of this cohort group is about the late 30s; after 20 years, their age is about the late 50s. The time span of the constructed longitudinal data gave us a pretty wide observation window to explore how age affects a faculty's productivity.</p> <p>Table 1 provides the descriptive statistics of our core sample and their characteristics in all covariates. In order to investigate how different types of faculty react to the promotion system, we further identify 84 faculty who are ranked at the top 40% of the most productive researchers in their own fields and label them as the early promise faculty. Our variables include time-invariant, time-variant and time-related ones. The time-invariant variables include gender and academic field. The time-variant variable includes faculty rank. Age, which necessarily changes with time, constitutes the time-related variable. We also control faculty members with or without administrative positions, their teaching hours in graduate and undergraduate school, and the number of advisees in master and PhD levels.</p> <p>Table 1. Descriptive statistics.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Group</td><td>Early promise</td><td>Non-early promise</td></tr><tr><td>Variables</td><td><italic>N</italic></td><td>Obs</td><td>Mean/%</td><td><italic>N</italic></td><td>Obs</td><td>Mean/%</td></tr></thead><tbody><tr><td>Gender</td><td /><td /><td /><td /><td /><td /></tr><tr><td> Male</td><td char=".">69</td><td char=".">901</td><td char=".">80%</td><td char=".">148</td><td char=".">2093</td><td char=".">72%</td></tr><tr><td> Female</td><td char=".">15</td><td char=".">226</td><td char=".">20%</td><td char=".">56</td><td char=".">819</td><td char=".">28%</td></tr><tr><td>Age</td><td char=".">84</td><td char=".">1127</td><td char=".">41.89</td><td char=".">208</td><td char=".">2912</td><td char=".">42.99</td></tr><tr><td>Faculty Rank</td><td /><td /><td /><td /><td /><td /></tr><tr><td> Assistant Professor</td><td char=".">46</td><td char=".">240</td><td char=".">21%</td><td char=".">97</td><td char=".">624</td><td char=".">21%</td></tr><tr><td> Associate Professor</td><td char=".">80</td><td char=".">538</td><td char=".">48%</td><td char=".">141</td><td char=".">1163</td><td char=".">40%</td></tr><tr><td> Professor</td><td char=".">52</td><td char=".">349</td><td char=".">31%</td><td char=".">118</td><td char=".">1125</td><td char=".">39%</td></tr><tr><td>Field</td><td /><td /><td /><td /><td /><td /></tr><tr><td> STEM</td><td char=".">29</td><td char=".">412</td><td char=".">37%</td><td char=".">82</td><td char=".">1249</td><td char=".">43%</td></tr><tr><td> Non-STEM</td><td char=".">55</td><td char=".">715</td><td char=".">63%</td><td char=".">122</td><td char=".">1663</td><td char=".">57%</td></tr><tr><td>Administration Position</td><td /><td /><td /><td /><td /><td /></tr><tr><td> Without</td><td char=".">83</td><td char=".">921</td><td char=".">82%</td><td char=".">204</td><td char=".">2471</td><td char=".">85%</td></tr><tr><td> With</td><td char=".">28</td><td char=".">206</td><td char=".">18%</td><td char=".">57</td><td char=".">441</td><td char=".">15%</td></tr><tr><td>Teaching hours</td><td /><td /><td /><td /><td /><td /></tr><tr><td> Grad School</td><td char=".">84</td><td char=".">1127</td><td char=".">6.42</td><td char=".">208</td><td char=".">2912</td><td char=".">5.75</td></tr><tr><td> Undergraduate School</td><td char=".">84</td><td char=".">1127</td><td char=".">8.36</td><td char=".">208</td><td char=".">2912</td><td char=".">8.98</td></tr><tr><td>Master advisees</td><td char=".">84</td><td char=".">1127</td><td char=".">4.39</td><td char=".">208</td><td char=".">2912</td><td char=".">3.67</td></tr><tr><td>PhD advisees</td><td char=".">84</td><td char=".">1127</td><td char=".">1.05</td><td char=".">208</td><td char=".">2912</td><td char=".">0.89</td></tr><tr><td>Granted Research Projects</td><td char=".">84</td><td char=".">1127</td><td char=".">1.51</td><td char=".">208</td><td char=".">2912</td><td char=".">1.07</td></tr></tbody></table> </ephtml> </p> <p>Table 2 provides summary statistics for the sample disaggregated into life course periods. Each row in the table corresponds to a five-year period in the life course. For each period, the column of individual indicates the number of individuals with experience at least equal to the minimum year in each range. The table lists the mean number of granted research projects per year, the variance, and the relative frequency of project counts for each period. Moving down the second and third columns of the table, we find that the mean number of granted projects is highest among the all periods followed by the fourth column decreasing sharply. The most common number of granted research projects for a given year is zero throughout the life course.</p> <p>Table 2. Statistics on the frequency of granted research project over the life course.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>Period</td><td>Individual</td><td>Mean</td><td>SD</td><td><italic>F</italic>(0)</td><td><italic>F</italic>(1)</td><td><italic>F</italic>(2)</td><td><italic>F</italic>(3)</td><td><italic>F</italic>(4)</td><td><italic>F</italic>(5+)</td></tr></thead><tbody><tr><td>1–5</td><td char=".">288</td><td char=".">1.17</td><td char=".">1.03</td><td char=".">0.20</td><td char=".">0.58</td><td char=".">0.14</td><td char=".">0.04</td><td char=".">0.02</td><td char=".">0.01</td></tr><tr><td>6–10</td><td char=".">284</td><td char=".">1.27</td><td char=".">1.22</td><td char=".">0.21</td><td char=".">0.51</td><td char=".">0.16</td><td char=".">0.06</td><td char=".">0.02</td><td char=".">0.02</td></tr><tr><td>11–15</td><td char=".">208</td><td char=".">1.23</td><td char=".">1.26</td><td char=".">0.26</td><td char=".">0.47</td><td char=".">0.14</td><td char=".">0.07</td><td char=".">0.03</td><td char=".">0.01</td></tr><tr><td>16–20</td><td char=".">121</td><td char=".">1.00</td><td char=".">1.37</td><td char=".">0.40</td><td char=".">0.41</td><td char=".">0.10</td><td char=".">0.03</td><td char=".">0.03</td><td char=".">0.01</td></tr></tbody></table> </ephtml> </p> <p>Note: Period is the length of employment.</p> <hd id="AN0153185467-9">Estimation strategy</hd> <p>We propose a quadratic growth curve model to fit our data with multilevel modeling of repeated measures. Three issues related to the possibility of awarded granted research projects should be addressed carefully. First, the budgetary amount that the NSF spends annually on sponsoring the university research varies. This could directly affect the possibility of faculty to receive a grant, even if the quality of the research proposal is equal to previous proposals. Second, the review decision is not only based on the quality of the research proposal, but also on the faculty's performance in previous years. This practice created a residual autocorrelation problem. Third, except for the change of the faculty's age, the teaching load and working environment could affect the faculty's performance in a positive or negative way. Failing to take this reality into consideration would bias our estimation.</p> <p>To overcome these issues, we modify our model in three ways. First, we add a centered year variable to control the unobserved time-trend, which comes from the change of the external environment (Baser and Pema [<reflink idref="bib1" id="ref42">1</reflink>]). Second, we extend the model to allow an AR(<reflink idref="bib1" id="ref43">1</reflink>) structure for the year-level residuals. Finally, we add other explanatory factors as control covariates. This practice can provide an accurate result for our estimation. The proposed random intercept model takes the form of equation:</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><msub><mi>Y</mi><mrow><mi>i</mi><mi>t</mi><mo>+</mo><mn>1</mn></mrow></msub><mo>=</mo><msub><mtext fontfamily="times">β</mtext><mrow><mn>0</mn><mi>t</mi></mrow></msub><mo>+</mo><msub><mtext fontfamily="times">β</mtext><mn>1</mn></msub><mi>a</mi><mi>g</mi><msub><mi>e</mi><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><msub><mtext fontfamily="times">β</mtext><mn>2</mn></msub><mi>a</mi><mi>g</mi><msubsup><mi>e</mi><mrow><mi>i</mi><mi>t</mi></mrow><mn>2</mn></msubsup><mo>+</mo><msub><mtext fontfamily="times">β</mtext><mn>3</mn></msub><msub><mi>X</mi><mrow><mi>i</mi><mi>t</mi></mrow></msub><mo>+</mo><msub><mtext fontfamily="times">β</mtext><mn>4</mn></msub><mi>c</mi><mi>e</mi><mi>n</mi><mi>t</mi><mi>e</mi><mi>r</mi><mi>e</mi><mi>d</mi><mi>y</mi><mi>e</mi><mi>a</mi><msub><mi>r</mi><mi>i</mi></msub><mo>+</mo><msub><mi>e</mi><mrow><mi>i</mi><mi>t</mi></mrow></msub></math> </ephtml> </p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><msub><mtext fontfamily="times">β</mtext><mrow><mn>0</mn><mi>t</mi></mrow></msub><mo>=</mo><msub><mtext fontfamily="times">β</mtext><mn>0</mn></msub><mo>+</mo><msub><mi>u</mi><mrow><mn>0</mn><mi>t</mi></mrow></msub></math> </ephtml> </p> <p>where</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><msub><mi>Y</mi><mrow><mi>i</mi><mi>t</mi><mo>+</mo><mn>1</mn></mrow></msub></math> </ephtml> is the amount of granted research project for individual faculty <emph>i</emph> in year <emph>t</emph> + 1;</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi>a</mi><mi>g</mi><msub><mi>e</mi><mrow><mi>i</mi><mi>t</mi></mrow></msub></math> </ephtml> is a continuous variable indicating the age of faculty <emph>i</emph> in year <emph>t</emph>;</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi>a</mi><mi>g</mi><msubsup><mi>e</mi><mrow><mi>i</mi><mi>t</mi></mrow><mn>2</mn></msubsup></math> </ephtml> is quadratic term of</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mi>a</mi><mi>g</mi><msub><mi>e</mi><mrow><mi>i</mi><mi>t</mi></mrow></msub></math> </ephtml> ;</p> <p>Graph</p> <p> <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><msub><mi>X</mi><mrow><mi>i</mi><mi>t</mi></mrow></msub><mspace width="thickmathspace" /></math> </ephtml> contains a set of covariates for faculty <emph>i</emph> in year <emph>t</emph>.</p> <p>In the following section, we first test the validity of the quadratic function between age and research productivity. Second, we compare the research profile of research productivity among varied comparison groups. Finally, we investigate how faculty react to promotion during their life course.</p> <hd id="AN0153185467-10">Empirical findings</hd> <p></p> <hd id="AN0153185467-11">Quadratic function in life course</hd> <p>In order to confirm which hypothesized model best fits with our research sample, we proposed three different age-research productivity relationships, linear and non-linear, for the statistical examination. The regression results in Table 3 present the association between age and the number of granted research projects in three models. Referring to model 1 and model 3, the coefficients of age-related variables are statistically insignificant. Model 2 shows that the association between research productivity and age is following the curve relationship of the quadratic equation. Furthermore, the chi-square test of Model 2 is statistically significant, indicating the validity of the proposed model. Our study documented the hypothesized positive but it diminishes the impact of age on research productivity even if the covariates have been controlled.</p> <p>Table 3. Conditional growth curve model.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>IV: Granted Research Project</td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td>Model 1</td><td>Model 2</td><td>Model 3</td></tr></thead><tbody><tr><td><p><graphic href="cshe_a_1723535_ilm0006.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><mi xmlns="">e</mi></math></p></td><td char=".">−0.020</td><td char=".">(0.012)</td><td char=".">0.152**</td><td char=".">(0.047)</td><td char=".">0.114</td><td char=".">(0.385)</td></tr><tr><td><p><graphic href="cshe_a_1723535_ilm0007.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><msup xmlns=""><mi>e</mi><mn>2</mn></msup></math></p></td><td char="." /><td char="." /><td char=".">−0.001***</td><td char=".">(0.000)</td><td char=".">−0.001</td><td char=".">(0.009)</td></tr><tr><td><p><graphic href="cshe_a_1723535_ilm0008.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><msup xmlns=""><mi>e</mi><mn>3</mn></msup></math></p></td><td char="." /><td char="." /><td char="." /><td char="." /><td char=".">−0.000</td><td char=".">(0.000)</td></tr><tr><td>Gender</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> Male</td><td char=".">0.204*</td><td char=".">(0.086)</td><td char=".">0.209*</td><td char=".">(0.087)</td><td char=".">0.209*</td><td char=".">(0.087)</td></tr><tr><td>Faculty Rank</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> Associate Prof</td><td char=".">0.125*</td><td char=".">(0.054)</td><td char=".">0.114*</td><td char=".">(0.054)</td><td char=".">0.114*</td><td char=".">(0.054)</td></tr><tr><td> Professor</td><td char=".">0.362***</td><td char=".">(0.072)</td><td char=".">0.359***</td><td char=".">(0.072)</td><td char=".">0.358***</td><td char=".">(0.073)</td></tr><tr><td>Fields</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> STEM</td><td char=".">0.363***</td><td char=".">(0.079)</td><td char=".">0.383***</td><td char=".">(0.079)</td><td char=".">0.383***</td><td char=".">(0.079)</td></tr><tr><td>Administration</td><td char=".">0.111*</td><td char=".">(0.050)</td><td char=".">0.103*</td><td char=".">(0.050)</td><td char=".">0.103*</td><td char=".">(0.050)</td></tr><tr><td>Teaching hours</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> Grad Level</td><td char=".">0.002</td><td char=".">(0.005)</td><td char=".">0.002</td><td char=".">(0.005)</td><td char=".">0.002</td><td char=".">(0.005)</td></tr><tr><td> College Level</td><td char=".">−0.002</td><td char=".">(0.004)</td><td char=".">−0.002</td><td char=".">(0.004)</td><td char=".">−0.002</td><td char=".">(0.004)</td></tr><tr><td>Master Advisees</td><td char=".">0.061***</td><td char=".">(0.007)</td><td char=".">0.057***</td><td char=".">(0.007)</td><td char=".">0.057***</td><td char=".">(0.007)</td></tr><tr><td>PhD Advisees</td><td char=".">0.077***</td><td char=".">(0.013)</td><td char=".">0.070***</td><td char=".">(0.013)</td><td char=".">0.070***</td><td char=".">(0.013)</td></tr><tr><td>Constant</td><td char=".">1.266*</td><td char=".">(0.510)</td><td char=".">−2.382*</td><td char=".">(1.091)</td><td char=".">−1.869</td><td char=".">(5.356)</td></tr><tr><td>Random intercept model</td></tr><tr><td /><td>Estimates</td><td>Std. Err.</td><td>Estimates</td><td>Std. Err.</td><td>Estimates</td><td>Std. Err.</td></tr><tr><td>Id: identity</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> var(_cons)</td><td>0.264***</td><td>(0.032)</td><td>0.268***</td><td>(0.032)</td><td>0.268***</td><td>(0.032)</td></tr><tr><td>Residual: AR(1)</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> rho</td><td>0.343***</td><td>(0.017)</td><td>0.343***</td><td>(0.017)</td><td>0.343***</td><td>(0.017)</td></tr><tr><td> var(e)</td><td>0.729***</td><td>(0.020)</td><td>0.726***</td><td>(0.020)</td><td>0.726***</td><td>(0.020)</td></tr><tr><td>Centered Year</td><td>Year</td><td>Year</td><td>Year</td></tr><tr><td><italic>N</italic> of observation</td><td>4039</td><td>4039</td><td>4039</td></tr><tr><td><italic>N</italic> of faculty</td><td>288</td><td>288</td><td>288</td></tr><tr><td>Wald <italic>χ</italic><sup>2</sup></td><td>298.80</td><td>311.04</td><td>310.99</td></tr><tr><td>Prob > <italic>χ</italic><sup>2</sup></td><td>0.000***</td><td>0.000***</td><td>0.000***</td></tr></tbody></table> </ephtml> </p> <p>Note: Standard errors in parentheses. *<emph>p</emph> <.05. **<emph>p</emph> <.01. ***<emph>p</emph> <.001.</p> <p>Besides the age variables, Model 2 also shows that male faculty who have a higher faculty rank, work in the STEM field or have more graduate advisees are more productive compared to their colleagues. At the same age, higher rank faculty are more productive compared to their lower rank counterparts. These findings are consistent with previous studies (Wanner, Lewis, and Gregorio [<reflink idref="bib23" id="ref44">23</reflink>]; Becher [<reflink idref="bib3" id="ref45">3</reflink>]; Stack [<reflink idref="bib20" id="ref46">20</reflink>]). However, one new finding emerges: having an administrative position is positively associated with the research productivity of faculty going against the findings of Goodwin and Sauer ([<reflink idref="bib8" id="ref47">8</reflink>]). One temporary explanation for such difference is the different contexts between American and Taiwan campuses. Though the utility maximizing theory also assumes 'shifting effort to non-research activities' (including administration) is generating greatest utility for elder faculty, within the Taiwanese context, the faculty who have better research capacity are more likely to be invited to take an administrative position. Additionally, the administrative position itself could benefit them by increasing their access to the institutional and social resources which could help their own research agenda. Such social-cultural variations between the Western and Eastern academic systems can explain why the utility maximizing theory is not so explanatory in Taiwan (Chang and Yang [<reflink idref="bib5" id="ref48">5</reflink>]).</p> <hd id="AN0153185467-12">Heterogeneity in changing productivity by unconditional model</hd> <p>Once the general model was confirmed, we began to investigate the heterogeneity effect of age by studying their life trajectory. Table 4 presents the coefficient of age based on the quadratic equation in different comparison combinations while Figure 2 visualizes their life-course pattern. Consistent with Cole's ([<reflink idref="bib6" id="ref49">6</reflink>]) and Bayer and Dutton's ([<reflink idref="bib2" id="ref50">2</reflink>]) studies, male and female faculty all reach their output peaks at around age 41. This observation also could be applied to STEM and Non-STEM faculty. It is noteworthy that even at age 55 when declining to the end of their later career, male faculty and STEM faculty's productivity are still superior to female and Non-STEM faculty who are at their peak.</p> <p>PHOTO (COLOR): Figure 2. Heterogeneity effect unconditional. Note: double–exponential smoothing is applied.</p> <p>Table 4. Unconditional growth curve model by comparison groups.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>IV: No. of Granted Project</td><td>Model 1</td><td>Model 2</td></tr><tr><td>(1)</td><td>(2)</td><td>(3)</td><td>(4)</td></tr><tr><td>Male</td><td>Female</td><td>Foreign</td><td>Domestic</td></tr></thead><tbody><tr><td><p><graphic href="cshe_a_1723535_ilm0009.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><mi xmlns="">e</mi></math></p></td><td>0.308*</td><td>(0.062)</td><td>0.252***</td><td>(0.059)</td><td char=".">0.259***</td><td char=".">(0.057)</td><td char=".">0.424***</td><td char=".">(0.091)</td></tr><tr><td><p><graphic href="cshe_a_1723535_ilm0010.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><msup xmlns=""><mi>e</mi><mn>2</mn></msup></math></p></td><td>−0.003***</td><td>(0.000)</td><td>−0.002***</td><td>(0.000)</td><td char=".">−0.002***</td><td char=".">(0.000)</td><td char=".">−0.004***</td><td char=".">(0.000)</td></tr><tr><td>Constant</td><td>−5.506***</td><td>(1.489)</td><td>−4.580***</td><td>(1.369)</td><td char=".">−4.758***</td><td char=".">(1.356)</td><td char=".">−8.002</td><td char=".">(2.380)</td></tr><tr><td>Centered Year</td><td>Yes</td><td>Yes</td><td>Yes</td><td>Yes</td></tr><tr><td><italic>N</italic> of obs.</td><td>2994</td><td>1045</td><td>3054</td><td>985</td></tr><tr><td><italic>N</italic> of faculty</td><td>217</td><td>71</td><td>218</td><td>70</td></tr><tr><td>Wald <italic>χ</italic><sup>2</sup></td><td>33.60</td><td>22.66</td><td>33.58</td><td>24.85</td></tr><tr><td>Prob > <italic>χ</italic><sup>2</sup></td><td>0.000***</td><td>0.000***</td><td>0.000***</td><td>0.000***</td></tr><tr><td /><td>Mode 3</td><td>Model 4</td></tr><tr><td>(5)</td><td>(6)</td><td>(7)</td><td>(8)</td></tr><tr><td>IV: No. of Granted Project</td><td>STEM</td><td>Non-STEM</td><td>Early Promise</td><td>Non-Early</td></tr><tr><td><p><graphic href="cshe_a_1723535_ilm0011.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><mi xmlns="">e</mi></math></p></td><td>0.340***</td><td>(0.090)</td><td>0.266***</td><td>(0.051)</td><td>0.389***</td><td>(0.109)</td><td>0.312***</td><td>(0.052)</td></tr><tr><td><p><graphic href="cshe_a_1723535_ilm0012.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><msup xmlns=""><mi>e</mi><mn>2</mn></msup></math></p></td><td>−0.003***</td><td>(0.000)</td><td>−0.003***</td><td>(0.000)</td><td>−0.004***</td><td>(0.001)</td><td>−0.003***</td><td>(0.000)</td></tr><tr><td>Constant</td><td>−6.078**</td><td>(2.194)</td><td>−4.777***</td><td>(1.191)</td><td>−7.100***</td><td>(2.511)</td><td>−6.397***</td><td>(1.298)</td></tr><tr><td>Centered Year</td><td>Yes</td><td>Yes</td><td>Yes</td><td>Yes</td></tr><tr><td><italic>N</italic> of obs<italic>.</italic></td><td>1661</td><td>2378</td><td>1127</td><td>2912</td></tr><tr><td><italic>N</italic> of faculty</td><td>111</td><td>177</td><td>84</td><td>204</td></tr><tr><td>Wald <italic>χ</italic><sup>2</sup></td><td>20.49</td><td>34.48</td><td>17.87</td><td>41.78</td></tr><tr><td>Prob > <italic>χ</italic><sup>2</sup></td><td>0.000***</td><td>0.000***</td><td>0.000***</td><td>0.000***</td></tr></tbody></table> </ephtml> </p> <p>Note: Standard errors in parentheses. *<emph>p</emph> <.05. **<emph>p</emph> <.01. ***<emph>p</emph> <.001</p> <p>Figure 2 also exposes a new finding which opposes to the popular stereotype that foreign-educated faculty are more research-capable. Our results show that domestic educated faculty perform better over their careers. The domestically educated faculty might not be very productive at the beginning of their career but they quickly catch up and even exceed the performance of their counterparts. By comparing the faculty who have a shining start to those who do not, we found that the cumulative advantage at the early stage is associated with the future performance. The early promising faculty take the lead throughout their whole career. When their counterparts had reached their peak at age of 43, the early promising faculty were still climbing. Their peak came later and their productivity over the whole career is much superior as well. This evidence directly supports the validity of cumulative advantage over the life trajectory.</p> <p>Similarly, as the obsolescence theory has argued, declining research productivity is an inevitable process that comes along with aging. Table 4 and Figure 2 confirm such prediction with much less publications after age 50 in every dimension. STEM subjects even suffer greater productivity setbacks compared to Non-STEM subject as result of rapid knowledge creation and replacement as we have previously argued. In the theoretical explanation, these hard scientists even face harsh publication competition if the mainstream paradigm were under fundamental revolution, making elder faculty intellectually incapable, or obsolescent.</p> <hd id="AN0153185467-13">Various productivity among comparison groups by conditional model</hd> <p>The unconditional growth curve model depicts the profile of research productivity among varied comparison groups. We further control the covariates and investigate the change of age's effect from the unconditional to the conditional model. The conditional estimation results in Table 5 show that although the impact of age decreases slightly, the superiority of the early promising and domestic faculty compared to their counterparts remains. Referring to Model 1 and Model 3, the marginal effect of age on the productivity in the group of the early promising faculty and domestic faculty are much higher than their comparison groups. This implies that given all the covariates are equal, at the same age, the early promising faculty and domestic faculty still engage in more research projects.</p> <p>Table 5. Conditional growth curve model by comparison groups.</p> <p> <ephtml> <table><thead valign="bottom"><tr><td>IV: Granted Research Project</td><td>Model 1</td><td>Model 2</td><td>Model 3</td></tr><tr><td>(1)</td><td>(2)</td><td>(3)</td><td>(4)</td><td>(5)</td><td>(6)</td></tr><tr><td>Early</td><td>Non-Early</td><td>STEM</td><td>Non-STEM</td><td>Foreign</td><td>Domestic</td></tr></thead><tbody><tr><td><p><graphic href="cshe_a_1723535_ilm0013.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><mi xmlns="">e</mi></math></p></td><td char=".">0.286** (0.105)</td><td char=".">0.157** (0.051)</td><td char=".">0.048 (0.086)</td><td char=".">0.194*** (0.051)</td><td char=".">0.128* (0.055)</td><td char=".">0.238** (0.089)</td></tr><tr><td><p><graphic href="cshe_a_1723535_ilm0014.gif" content-type="Graph" /><math xmlns="http://www.w3.org/1998/Math/MathML"><mi xmlns="">a</mi><mi xmlns="">g</mi><msup xmlns=""><mi>e</mi><mn>2</mn></msup></math></p></td><td char=".">−0.003** (0.001)</td><td char=".">−0.001*** (0.000)</td><td char=".">−0.000 (0.000)</td><td char=".">−0.002*** (0.000)</td><td char=".">−0.001** (0.000)</td><td char=".">−0.002** (0.000)</td></tr><tr><td>Gender</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> Male</td><td char=".">0.047 (0.180)</td><td char=".">0.163<sup>+</sup> (0.092)</td><td char=".">0.380<sup>+</sup> (0.211)</td><td char=".">0.152* (0.076)</td><td char=".">0.153 (0.097)</td><td char=".">0.367<sup>+</sup> (0.198)</td></tr><tr><td>Faculty Rank</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> Associate Prof</td><td char=".">0.008 (0.111)</td><td char=".">0.141* (0.060)</td><td char=".">0.162 (0.101)</td><td char=".">0.125* (0.058)</td><td char=".">0.128* (0.062)</td><td char=".">−0.024 (0.105)</td></tr><tr><td> Professor</td><td char=".">0.162 (0.157)</td><td char=".">0.477*** (0.079)</td><td char=".">0.518*** (0.137)</td><td char=".">0.315*** (0.075)</td><td char=".">0.422*** (0.083)</td><td char=".">0.037 (0.147)</td></tr><tr><td>Fields</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> STEM</td><td char=".">0.608*** (0.158)</td><td char=".">0.363*** (0.085)</td><td char="." /><td char="." /><td char=".">0.372*** (0.089)</td><td char=".">0.415* (0.178)</td></tr><tr><td>Administration</td><td char=".">−0.000 (0.106)</td><td char=".">0.140* (0.056)</td><td char=".">0.277** (0.102)</td><td char=".">0.053 (0.053)</td><td char=".">0.137* (0.059)</td><td char=".">−0.042 (0.093)</td></tr><tr><td>Teaching hours</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> Grad Level</td><td char=".">−0.007 (0.011)</td><td char=".">0.005 (0.006)</td><td char=".">−0.026* (0.011)</td><td char=".">0.014* (0.005)</td><td char=".">0.003 (0.006)</td><td char=".">−0.004 (0.010)</td></tr><tr><td> College Level</td><td char=".">0.004 (0.008)</td><td char=".">−0.003 (0.004)</td><td char=".">−0.021** (0.008)</td><td char=".">0.002 (0.004)</td><td char=".">−0.004 (0.004)</td><td char=".">−0.001 (0.008)</td></tr><tr><td>Master Advisees</td><td char=".">0.0536*** (0.014)</td><td char=".">0.056*** (0.008)</td><td char=".">0.080*** (0.011)</td><td char=".">0.034*** (0.008)</td><td char=".">0.053*** (0.008)</td><td char=".">0.065*** (0.013)</td></tr><tr><td>PhD Advisees</td><td char=".">0.085** (0.027)</td><td char=".">0.063*** (0.014)</td><td char=".">0.126*** (0.021)</td><td char=".">0.004 (0.016)</td><td char=".">0.049** (0.015)</td><td char=".">0.135*** (0.026)</td></tr><tr><td>Constant</td><td char=".">−4.901* (2.329)</td><td char=".">−2.925* (1.200)</td><td char=".">0.064 (2.012)</td><td char=".">−3.236** (1.175)</td><td char=".">−1.845 (1.282)</td><td char=".">−4.386* (2.125)</td></tr><tr><td /><td>random intercept model</td></tr><tr><td>Id: identity</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> var(_con)</td><td char=".">0.213*** (0.058)</td><td char=".">0.226*** (0.033)</td><td char=".">0.341*** (0.068)</td><td char=".">0.148*** (0.025)</td><td char=".">0.243*** (0.035)</td><td char=".">0.350*** (0.080)</td></tr><tr><td>Residual: AR(1)</td><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /><td char="." /></tr><tr><td> rho</td><td char=".">0.397*** (0.341)</td><td char=".">0.308*** (0.021)</td><td char=".">0.371*** (0.027)</td><td char=".">0.294*** (0.023)</td><td char=".">0.391*** (0.020)</td><td char=".">0.187*** (0.036)</td></tr><tr><td> var(e)</td><td char=".">0.970*** (0.055)</td><td char=".">0.627*** (0.020)</td><td char=".">0.951*** (0.043)</td><td char=".">0.549*** (0.019)</td><td char=".">0.720*** (0.025)</td><td char=".">0.723*** (0.037)</td></tr><tr><td>Centered Year</td><td char=".">Yes</td><td char=".">Yes</td><td char=".">Yes</td><td char=".">Yes</td><td char=".">Yes</td><td char=".">Yes</td></tr><tr><td><italic>N</italic> of observation</td><td char=".">1127</td><td char=".">2912</td><td char=".">1661</td><td char=".">2378</td><td char=".">3054</td><td char=".">985</td></tr><tr><td><italic>N</italic> of faculty</td><td char=".">84</td><td char=".">204</td><td char=".">111</td><td char=".">177</td><td char=".">218</td><td char=".">70</td></tr><tr><td>Wald <italic>χ</italic><sup>2</sup></td><td char=".">91.35</td><td char=".">273.07</td><td char=".">185.83</td><td char=".">94.09</td><td char=".">208.20</td><td char=".">117.43</td></tr><tr><td>Prob > <italic>χ</italic><sup>2</sup></td><td char=".">0.000***</td><td char=".">0.000***</td><td char=".">0.000***</td><td char=".">0.000***</td><td char=".">0.000***</td><td char=".">0.000***</td></tr></tbody></table> </ephtml> </p> <p>Note: Standard errors in parentheses. <sups>+</sups><emph>p</emph> <.10. *<emph>p</emph> <.05. **<emph>p</emph> <.01. ***<emph>p</emph> <.001</p> <p>Nonetheless, the superiority of STEM faculty observed in the unconditional model disappears when the covariates were taken into consideration. Having PhD advisees or an administrative position are two covariates in which the coefficients are significantly positive for STEM faculty but not for Non-STEM faculty. A simple <emph>t</emph>-test was applied to investigate the difference between STEM and Non-STEM faculty on these two covariates. The results show that STEM faculty have significantly more PhD advisees compared to their counterparts (<emph>t</emph>-value = 12.4771, degree of freedom = 4.037). PhD students are valuable manpower to support the operation of research projects. This scenario implies that within the group of STEM faculty, the number of PhD advisees is a more powerful predictor to their productivity instead of the age of faculty as observed in Non-STEM faculty group. Put differently, their productivity could be less vulnerable to the aging effect if they have the support of PhD students. Conversely, Non-STEM faculty are more reliant upon personal effort and capacity for productivity.</p> <p>Previously, we already indicated that having an administrative position in Taiwan has different interpretations under the utility maximizing theory. In Table 5, we again find out that non-early promising faculty, STEM faculty, and foreign degree holders with administrative jobs have better research productivity. These phenomena are against original theoretical assumptions. In fact, these faculties are also better off in research production due to the greater institutional and social resources. This also illustrates the different academic ecology with respect to the administrative position appointment from Western systems.</p> <hd id="AN0153185467-14">Faculty's reaction to the promotion system: 'reinforcement' matters</hd> <p>During the life course of the faculty's research profile, several empirical studies argued that the promotion system, as a stimulus or reinforcement, plays a key role to motivate their behaviors (Tien and Blackburn [<reflink idref="bib21" id="ref51">21</reflink>]). Ideally, we should randomly assign a group of faculty to the control group and another group of faculty to the treatment group. Because it is impossible to realize such experimental design, we obviate this technical issue by observing how two separate groups of faculty with apparently different characteristics react to the promotion during their life course. These two groups are the early promising group and the non-early promising group. Among 84 faculty, 13 in early promise group remain at the same faculty rank, 48 experienced transition of upper promotion once, and 23 twice. Among 204 faculty, 87 in the non-early promising group remain at the same faculty rank, 82 experienced transition of upper promotion once, and 35 twice. Since we do not have those faculty who had failed to get promotion, we limit our observation to those who have been promoted.</p> <p>Figure 3 displays the cumulative percentage distribution of faculty with at least one granted research project before and after their transition to the upper faculty rank. The threshold of 0 indicated by vertical lines refers to the year of transition. The blue line indicates the trajectory of the early promising faculty with grants; while the red line indicates the non-early promising faculty. For the early promising group, the number records of faculty with one to three years before and after promotion from assistant to associate professor is 42, 38 and 32 correspondingly. The number records of faculty in the non-early promising group is 85, 81 and 72. For the early promising group, the number records of faculty with one to three years before and after promotion from associate to professor is 52, 49 and 38 correspondingly. The number records of faculty in the non-early promising group is 67, 60 and 51.</p> <p>PHOTO (COLOR): Figure 3. Schedule of reinforcement.</p> <p>During the 3 years before and after the promotion from assistant professor to associate professor, the percentage of active faculty in the early promising group remains very stable. In contrast, the non-early promising faculty present an apparently hump-shaped pattern. The fluctuation during the transition implies that the non-early promising faculty are more sensitive to the intermittent schedule of reinforcement. The behavior change pattern is also dramatic while observing the promotion from associate professor to professor. Before promotion, two groups shared a common pattern both on the percentage and the trend. However, after promotion the early promising faculty experienced no change, not even slight climbs. On the contrary, the active percentage of the non-early promising faculty declines sharply. This difference further confirms that the early promising faculty tend to keep their pace and remain active in research. The intermittent schedule of reinforcement has an apparent impact on the non-early promising faculty. This finding is opposite to the results presented by Tien and Blackburn ([<reflink idref="bib21" id="ref52">21</reflink>]), who argued that there is no enough evidence to support the effect of reinforcement. Such differences probably can be attributed to the less research focus two decades ago in Taiwan. Moreover, the academic promotion system before was entirely different from the current ones. These nuanced variations might explain the different outcomes.</p> <p>In the previous section, we have confirmed that the early promising group have superior performance over the whole life course compared to non-early promising ones. The Matthew effect, as argued by the cumulative advantage theory, could be cited to provide a theoretical explanation thereby gaining substantial support from our empirical findings. For those non-early promising faculty, they are really driven by the external 'reinforcement' or the promotion system. Instead, the early promising group tends to be highly and intrinsically motivated, which makes them less sensitive to the external reinforcement. Therefore, the mechanisms for supporting and maintaining research output from a life-course perspective becomes a critical managerial and policy issue for university leaders and decision-makers in government.</p> <p>There are three major novelties of this research. First, domestic PhD degree holders surpassed their colleagues with foreign degrees at around 40 years old. This finding requires further investigation as to how these locally trained PhDs improve their performance along with the maturing of their research career. The second novelty is associated with the better research productivity of administrative position holders. As argued previously, this phenomenon relates to the social-cultural contexts within the Taiwanese academic community. At the same time, the utility maximizing theory might help to explain this result as having an administrative job may act as a strategy helping faculty to lift their research outputs. Finally, the obsolescence theory successfully predicts the lowering academic productivity that comes with aging. Our direct findings suggest different speeds of decline for STEM and non-STEM faculty. This interesting result might be worthwhile for further exploration.</p> <hd id="AN0153185467-15">Conclusion</hd> <p>Our study confirms that age is a key factor shaping research productivity from a life-course perspective. We also confirm the existence of the Matthew effect in research productivity of faculty, where early success has an association with the later career path. These faculty members tend to be more self-motivated researchers. Additionally, reinforcement as the core notion of the 'cumulative advantage theory' also gains substantial support for non-early promising faculty who are significantly affected by the promotion system. Timing and appropriate and effective reinforcement mechanisms, therefore, should play essential roles in driving and maintaining research productivity particularly for them. Otherwise, their research outputs will decline dramatically.</p> <p>As to the utility maximizing theory, having an administrative position in the Taiwanese academic community tends to be more productive, contrary to the theoretical assumption. This issue deserves to be explored further through an academic cultural perspective. Finally, the obsolescence theory gains wide support from empirical results. Elder faculty with much lower productivity really reflect the 'decline of intellectual ability'. Moreover, STEM faculty experiencing greater decreases in research output than non-STEM ones coincide with the reality of rapid replacement of knowledge in the hard sciences.</p> <p>Although we have provided solid evidence to the theoretical query through our sampling strategy, our findings are rooted in the local context and have limited generality. Over the past 20 years, the popularity of world-class university policy has been changing the culture of scientific production in the higher education system (Fu, Baker, and Zhang [<reflink idref="bib7" id="ref53">7</reflink>]). Such external shocks could have impacted on how the younger generation of faculty commit to research. Other researchers could consider to add new cohort groups and compare their research profile with their elderly peers. This comparison could further extend our knowledge about the impact of the age effect in different periods.</p> <p>Based on earlier findings, we identify several significant policy and managerial implications. First of all, the sampled university, as a representative of Taiwanese research-oriented universities, faced an aging faculty and subsequently a declining of research productivity as a whole. Thus, it is critical to ensure a stable faculty replacement mechanism among different age generations. Secondly, according to the cumulative advantage theory, it is crucial to invest resources and establish supporting mechanisms for young faculty who are at the stage of taking-off. Not only can their academic reputation be well-established, but their research ability can be further cultivated soon after their post-doctoral stage. Fortunately, since 2017 the Taiwanese Ministry of Science and Technology has paid more attention to cultivating talented young scholars through special funding projects. Additionally, young scholars who show research potential/outcomes could also benefit from additional salary sponsored by the Sprout Project (深耕計畫) of the Ministry of Education. Thirdly, a merit-based mechanism or reward structure acting as a schedule of reinforcement is required to stimulate the research momentum of the non-early promising faculty as they are extrinsically driven for productivity. Finally, it is our suggestion that administrators should also be the academic leader within the university for greater and better performance. Faculty with administrative positions can be the 'engine' driving more research productivity.</p> <hd id="AN0153185467-16">Disclosure statement</hd> <p>No potential conflict of interest was reported by the author(s).</p> <ref id="AN0153185467-17"> <title> References </title> <blist> <bibl id="bib1" idref="ref42" type="bt">1</bibl> <bibtext> Baser, O., and E. Pema. 2004. " Publications Over the Academic Life-course: Evidence for Academic Economists." Economics Bulletin 1 : 1 – 8.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref4" type="bt">2</bibl> <bibtext> Bayer, A. E., and J. E. Dutton. 1977. " Career age and Research-professional Activities of Academic Scientists: Tests of Alternative Nonlinear Models and Some Implications for Higher Education Faculty Policies." The Journal of Higher Education 48 (3): 259 – 82.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref15" type="bt">3</bibl> <bibtext> Becher, T. 1994. " The Significance of Disciplinary Differences." Studies in Higher Education 19 (2): 151 – 61. doi: 10.1080/03075079412331382007</bibtext> </blist> <blist> <bibl id="bib4" idref="ref1" type="bt">4</bibl> <bibtext> Chan, S. J. 2015. " Massification of Higher Education in Taiwan: Shifting Pressure from Admission to Employment." Higher Education Policy 28 (1): 17 – 33. doi: 10.1057/hep.2014.33</bibtext> </blist> <blist> <bibl id="bib5" idref="ref48" type="bt">5</bibl> <bibtext> Chang, S. J., and C. C. Yang. 2017. " Hybrid University in Taiwan: The Prominence of Traditional Intellects." Studies in Higher Education 42 (10): 1853 – 69. doi: 10.1080/03075079.2017.1376869</bibtext> </blist> <blist> <bibl id="bib6" idref="ref24" type="bt">6</bibl> <bibtext> Cole, S. 1979. " Age and Scientific Performance." American Journal of Sociology 84 (4): 958 – 77. doi: 10.1086/226868</bibtext> </blist> <blist> <bibl id="bib7" idref="ref53" type="bt">7</bibl> <bibtext> Fu, Y. C., D. P. Baker, and L. Zhang. 2018. " Engineering a World Class University? The Impact of Taiwan's World Class University Project on Scientific Productivity." Higher Education Policy, 1 – 16. doi: 10.1057/s41307-018-0110-z</bibtext> </blist> <blist> <bibl id="bib8" idref="ref11" type="bt">8</bibl> <bibtext> Goodwin, T. H., and R. D. Sauer. 1995. " Life Course Productivity in Academic Research: Evidence from Cumulative Publication Histories of Academic Economists." Southern Economic Journal 61 : 728 – 43. doi: 10.2307/1060993</bibtext> </blist> <blist> <bibl id="bib9" idref="ref30" type="bt">9</bibl> <bibtext> Holliday, E. B., R. Jagsi, L. D. Wilson, M. Choi, C. R. Thomas Jr, and C. D. Fuller. 2014. " Gender Differences in Publication Productivity, Academic Position, Career Duration and Funding among US Academic Radiation Oncology Faculty." Academic Medicine 89 (5): 767 – 73. doi: 10.1097/ACM.0000000000000229</bibtext> </blist> <blist> <bibtext> Hu, Q., and T. G. Gill. 2000. " IS Faculty Research Productivity: Influential Factors and Implications." Information Resources Management Journal 13 (2): 15 – 25. doi: 10.4018/irmj.2000040102</bibtext> </blist> <blist> <bibtext> Kwiek, M. 2018. Changing European Academics: A Comparative Study of Social Stratification, Work Patterns and Research Productivity. New York : Routledge.</bibtext> </blist> <blist> <bibtext> Kyvik, S. 1990. " Age and Scientific Productivity. Differences Between Fields of Learning." Higher Education 19 (1): 37 – 55. doi: 10.1007/BF00142022</bibtext> </blist> <blist> <bibtext> Levin, S. G., P. E. Stephan. 1984. " Methodological Issues Encountered in Estimating the Relationship Between Publishing and Age." In George Sarton Centennial, edited by W. Callebaut. Ghent : Communication & Cognition.</bibtext> </blist> <blist> <bibtext> Levin, S. G., and P. E. Stephan. 1989. " Age and Research Productivity of Academic Scientists." Research in Higher Education 30 (5): 531 – 49. doi: 10.1007/BF00992202</bibtext> </blist> <blist> <bibtext> Merton, R. K. 1968. " The Matthew Effect in Science: The Reward and Communication Systems of Science are Considered." Science 159 (3810): 56 – 63. doi: 10.1126/science.159.3810.56</bibtext> </blist> <blist> <bibtext> Perry, R. P., R. A. Clifton, V. H. Menec, C. W. Struthers, and R. J. Menges. 2000. " Faculty in Transition: A Longitudinal Analysis of Perceived Control and Type of Institution in the Research Productivity of Newly Hired Faculty." Research in Higher Education 41 (2): 165 – 94. doi: 10.1023/A:1007091104399</bibtext> </blist> <blist> <bibtext> Rauber, M., and H. W. Ursprung. 2008. " Life Cycle and Cohort Productivity in Economic Research: The Case of Germany." German Economic Review 9 (4): 431 – 56. doi: 10.1111/j.1468-0475.2008.00447.x</bibtext> </blist> <blist> <bibtext> Sabharwal, M. 2013. " Comparing Research Productivity Across Disciplines and Career Stages." Journal of Comparative Policy Analysis: Research and Practice 15 (2): 141 – 63.</bibtext> </blist> <blist> <bibtext> Scimago Institutions Rankings. 2019. SJR Scimago Journal & Country Rank. Retrieved from https://<ulink href="http://www.scimagojr.com/countryrank.php">www.scimagojr.com/countryrank.php</ulink>.</bibtext> </blist> <blist> <bibtext> Stack, S. 2004. " Gender, Children and Research Productivity." Research in Higher Education 45 : 891 – 920. doi: 10.1007/s11162-004-5953-z</bibtext> </blist> <blist> <bibtext> Tien, F. F., and R. T. Blackburn. 1996. " Faculty Rank System, Research Motivation, and Faculty Research Productivity: Measure Refinement and Theory Testing." The Journal of Higher Education 67 (1): 2 – 22. doi: 10.2307/2943901</bibtext> </blist> <blist> <bibtext> Tuckman, H. P. 1976. Publication, Teaching, and the Academic Reward Structure. New York : Lexington Books.</bibtext> </blist> <blist> <bibtext> Wanner, R. A., L. S. Lewis, and D. I. Gregorio. 1981. " Research Productivity in Academia: A Comparative Study of the Sciences, Social Sciences and Humanities." Sociology of Education 54 (4): 238 – 53. doi: 10.2307/2112566</bibtext> </blist> </ref> <aug> <p>By Yuan-Chih Fu; Sheng-Ju Chan; Shi-Ming Huang and Ya-Hui Lee</p> <p>Reported by Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib19" firstref="ref2"></nolink> <nolink nlid="nl2" bibid="bib14" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib17" firstref="ref6"></nolink> <nolink nlid="nl4" bibid="bib18" firstref="ref7"></nolink> <nolink nlid="nl5" bibid="bib21" firstref="ref8"></nolink> <nolink nlid="nl6" bibid="bib10" firstref="ref9"></nolink> <nolink nlid="nl7" bibid="bib16" firstref="ref13"></nolink> <nolink nlid="nl8" bibid="bib20" firstref="ref16"></nolink> <nolink nlid="nl9" bibid="bib23" firstref="ref17"></nolink> <nolink nlid="nl10" bibid="bib15" firstref="ref18"></nolink> <nolink nlid="nl11" bibid="bib12" firstref="ref19"></nolink> <nolink nlid="nl12" bibid="bib40" firstref="ref20"></nolink> <nolink nlid="nl13" bibid="bib11" firstref="ref21"></nolink> <nolink nlid="nl14" bibid="bib13" firstref="ref23"></nolink> <nolink nlid="nl15" bibid="bib22" firstref="ref36"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1316346
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Life Course Productivity Model to Analyze Academic Research Issues: A Longitudinal Analysis at One Taiwanese University
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Fu%2C+Yuan-Chih%22">Fu, Yuan-Chih</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0003-1397-2207">0000-0003-1397-2207</externalLink>)<br /><searchLink fieldCode="AR" term="%22Chan%2C+Sheng-Ju%22">Chan, Sheng-Ju</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0003-3830-5304">0000-0003-3830-5304</externalLink>)<br /><searchLink fieldCode="AR" term="%22Huang%2C+Shi-Ming%22">Huang, Shi-Ming</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0001-6062-7153">0000-0001-6062-7153</externalLink>)<br /><searchLink fieldCode="AR" term="%22Lee%2C+Ya-Hui%22">Lee, Ya-Hui</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0001-5802-9146">0000-0001-5802-9146</externalLink>)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Studies+in+Higher+Education%22"><i>Studies in Higher Education</i></searchLink>. 2021 46(11):2491-2505.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 15
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2021
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Higher+Education%22">Higher Education</searchLink><br /><searchLink fieldCode="EL" term="%22Postsecondary+Education%22">Postsecondary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Research+Universities%22">Research Universities</searchLink><br /><searchLink fieldCode="DE" term="%22College+Faculty%22">College Faculty</searchLink><br /><searchLink fieldCode="DE" term="%22Teacher+Researchers%22">Teacher Researchers</searchLink><br /><searchLink fieldCode="DE" term="%22Productivity%22">Productivity</searchLink><br /><searchLink fieldCode="DE" term="%22Institutional+Research%22">Institutional Research</searchLink><br /><searchLink fieldCode="DE" term="%22Reinforcement%22">Reinforcement</searchLink><br /><searchLink fieldCode="DE" term="%22Noninstructional+Responsibility%22">Noninstructional Responsibility</searchLink><br /><searchLink fieldCode="DE" term="%22Age%22">Age</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Taiwan%22">Taiwan</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/03075079.2020.1723535
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0307-5079
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Research productivity has been a critical issue in terms of academic development in higher education. In this study, we adopt a life-course perspective to examine the personal factors, mostly age-related, affecting research productivity in a Taiwanese research-oriented university. Covering a time series of 20 years, our dataset includes individual research performance of faculty and other relevant covariates over their life course. The growth curve model designed for multilevel modeling of repeated measures is applied to capture the age effect. Our analysis contributes to the thread of this literature in several dimensions. First, the faculty's early academic achievement is positively associated with their later performance providing support for the cumulative advantage theory. Unlike the prediction of the utility maximizing theory, faculty with an administrative position leads to higher productivity. Finally, reinforcement still plays a critical role in regulating the productivity for non-early promising faculty.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2021
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1316346
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1316346
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/03075079.2020.1723535
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 2491
    Subjects:
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Research Universities
        Type: general
      – SubjectFull: College Faculty
        Type: general
      – SubjectFull: Teacher Researchers
        Type: general
      – SubjectFull: Productivity
        Type: general
      – SubjectFull: Institutional Research
        Type: general
      – SubjectFull: Reinforcement
        Type: general
      – SubjectFull: Noninstructional Responsibility
        Type: general
      – SubjectFull: Age
        Type: general
      – SubjectFull: Taiwan
        Type: general
    Titles:
      – TitleFull: Life Course Productivity Model to Analyze Academic Research Issues: A Longitudinal Analysis at One Taiwanese University
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Fu, Yuan-Chih
      – PersonEntity:
          Name:
            NameFull: Chan, Sheng-Ju
      – PersonEntity:
          Name:
            NameFull: Huang, Shi-Ming
      – PersonEntity:
          Name:
            NameFull: Lee, Ya-Hui
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2021
          Identifiers:
            – Type: issn-print
              Value: 0307-5079
          Numbering:
            – Type: volume
              Value: 46
            – Type: issue
              Value: 11
          Titles:
            – TitleFull: Studies in Higher Education
              Type: main
ResultId 1