Question

According to the data in table 1, Chemistry Journals cost 3 times more than Agriculture Journals. Why would there be such a large disparity in these prices?

Information

Fracking the ecosystem: journal prices continue to rise, while resources remain flat, challenging all players to eke out the most value for students and scholars

Library Journal

Fracking is the process of injecting liquids or solids into rock under high pressure, releasing oil and gas deposits from fields that may have been previously thought to be of marginal quality or depleted to the point that it wouldn't be cost effective to develop using traditional methods alone. While a common practice to improve oil field productivity, consequences ranging from water contamination to earthquakes have been attributed to the practice.

What does fracking have to do with scholarly publishing and journal pricing? While the library financial landscape has improved since the depth of the Great Recession of 2007 to 2009, it still cannot be considered robust. As articles such as this one chronicle annual serials price increases, libraries, publishers, and vendors search for innovative ways to fulfill information needs within the finite, predefined budget environment. New business and access models ranging from the initial e-journal big deal packages, article pay per view, open access, mega-journals, and publisher e-journal database pricing have evolved in response to the environment; libraries, publishers, and vendors have merged, consolidated, or disappeared along the way. Just as fracking keeps the oil and gas flowing, these strategies enable the current scholarly publishing ecosystem to extract the necessary resources--intellectual and financial--to survive.

Stagnant economy

The U.S. economy continues to expand, and the public sector is experiencing positive growth, but that growth is not yet sufficient to return to pre-recession (2008) levels when adjusted for inflation. According to reports from the National Association of State Budget Officers (NASBO), overall state spending has increased compared with 2008, rising from $687 billion to an expenditure of $759 billion in 2015. However, when inflation is calculated based on the Consumer Price Index (CPI), the 2015 expenditures fall 4% short of the $789 billion that would be needed just to keep up. State general fund spending is expected to top $790 billion in 2016, but, again, that will be just shy of what would be needed to match 2008 levels. According to the NASBO report, "State budgets continue to grow at a moderate pace after several years of slow recovery in the national economy following the Great Recession.... This growth rate falls below the historical average, though the current inflation rate remains low as well."

Funding for K-12 education has seen some relief as a total of 41 states increased allotments.

Slow growth in public funding is reflected in budgets for higher education and libraries. A 2015 report from the Center on Budget and Policy Priorities notes that 47 states still spend less per student in higher education than was spent in 2007-08. In 2016, Inside Higher Ed reported this mixed economic message. Appropriations increased 4.1% for 2016, but this figure does not include Illinois and Pennsylvania, which, when reported, could pull down the percentage. Some 15 states are still appropriating less than they were five years ago.

A survey conducted by the Publishers Communication Group in 2015 indicated that growth in library funding was expected to be about 1.2% in 2015, not even enough to cover basic inflation let alone increases common to library materials. The strength of the dollar relative to other world currencies will add additional pressure for libraries, especially those outside of the United States, greatly magnifying the impact of annual price increases and slow-growth budgets.

Maximizing value

All players in the scholarly communications ecosystem, including publishers, libraries, and vendors, are going to have to develop creative ways to "frack" the system to squeeze added value out of a stagnant resource base. Slow budget growth and price inflation will force libraries to manage actively the resources they make available to their users in a number of ways. Decisions will be made using the best data available. Content needed by scholars will be provided by the most cost-effective means. Some libraries will be dropping big deals. Delivery of content will be a focus. Content does not need to be owned/collected to be useful. Budgets will be aligned to support current information needs within the constraints of available funds. It will be necessary to work with the broader community to build cooperative approaches to acquiring and preserving the scholarly record comprehensively.

Price, value, and expenditure

Price, value, and expenditure are all related, but each is measured in different ways. The value of content is an important factor to consider when making purchasing decisions, but the price on the invoice is what impacts the budget, and there needs to be reliable information on changes in price to assist with budget planning.

This price survey, with the exception of Table 3, uses a print-preferred pricing model based on the standard retail price for the titles in the selected indexes. Print pricing is used for consistency because not all publishers make their online-only pricing available, or they don't have a standard online-only retail price. However the index does contain some pricing for print + online and online (29% of the titles) but only if those were the only rates offered.

While many of the charts that accompany this article use the print pricing of journals, as noted in current and past price projections features, it's important to keep in mind that such a large portion of library budgets is now dedicated to the purchase of e-journal packages. Accordingly, we have measured the 2015 price increases of more than 5,000 e-journal packages handled by EBSCO. Our analysis indicates an average e-journal package price increase of 5.8% to 6.3%, down slightly from last year's average of 6.6%.

Expanding the data set

Traditionally, this survey has used the titles indexed in the ISI Arts and Humanities Index, Science Citation Abstracts, and Social Science Citation Index. We have presented the data from these indexes individually as well as in merged format, sorted by discipline and country of origin.

Historically, we have also analyzed the pricing of the full-text titles in EBSCO's Academic Search Premier database to represent the titles most commonly held in general academic and public library collections. Starting with this survey, we will be using the titles indexed by Academic Search Complete, which is now the most commonly found in its respective markets. This shift also makes the data more consistent with the ISI indexes as both sets of data are based on indexed titles, not on index titles in one set of data and full-text titles in another. The result of these changes is that we are now examining 7,124 priced titles indexed in Academic Search Complete. Last year the data set was 3,151 titles. (For an additional table analyzing 1,266 priced titles from Masterfile Complete, see the online version of this article at ow.ly/106fvp.)

This year we were also able to expand the study to include approximately 15,000 titles indexed by SCOPUS. Subsequently, this year's feature examines pricing for 18,473 unique titles, our largest sample to date. Increasing the sample makes the results more reliable, and like nearly all other data sources employed for this article, the increase in serial prices documented in the SCOPUS Table is again 6% for 2016.

2016 by the numbers

Tables 1-3 accompanying this article analyze average subscription prices by LC subject area and country of origin.

TABLE 1 (page 32) Average prices for science, technology, and medical (STM) serials remain the highest, compared with prices for serials in other subject areas. Chemistry has historically seen the highest average serial prices and that has not changed. This year's data reports the average price for chemistry journals is $5,105 annually. There were only slight shifts in relative rankings for subject areas based on the overall average prices for serials. Food science and astronomy swapped places as well as technology and zoology. As noted in the analysis of the charts, content from the five major publishers--Elsevier, Springer, Wiley, Taylor & Francis, and SAGE--continues to represent more than half (52%) of the titles in the merged ISI indexes.

TABLE 2 (page 33) There was little change in the relative order for the average price per title sorted by country of origin. Since so many serials are now priced in U.S. dollars, the exchange rate issues experienced by foreign libraries are not seen in this table. While the average price per title increased from previous years, titles from Russia and Ireland continue to have the highest cost per title for all foreign titles included in the merged ISI indexes. Hungary, Austria, the Netherlands, Singapore, Germany, Switzerland, England, and New Zealand round out the top ten countries with the highest cost per title in 2016.

TABLE 3 (page 33) Table 3 examines the titles in the combined ISI Arts and Humanities, Science Citation, and Social Sciences Citation indexes, which offer published rates for online formats. This table now has 61% of the titles in the full ISI-based tables. As in past years, the data reflects online only, print plus free online, and the first tier of any tiered pricing, with the common element being pricing for the online format. As in Table 1, STM titles are the most expensive, with chemistry titles having the highest average cost. The 2016 average cost for this set of titles is $1,486. While the 5% increase over last's year's average price of $1,413 may seem modest, it is important to bear in mind that the individual titles in the data set and associated pricing models reflected by Table 3 change each year so price increases can be due to changes in the data as well as actual price changes. What does remain consistent is the annual increase falling in the 5% to 6% range and the relative high cost of titles by subject areas, with science and technology areas at the top of the chart. The average costs are lower in Table 3 than in the ISI-based tables that include print. This is owing to some large, high-price publishers not having retail prices for online versions of their journals.

The 2017 forecast

The 2017 serials marketplace will continue to see steady increases in prices, with no indicators to the contrary. Those involved with the scholarly communications marketplace are all fracking the system, trying to release more value from a limited resource base. Drivers continue to include increased research outputs, budget compression, price inflation, the need to fund new research and education directions, open access, government mandates, new assessment and evaluation tools, and changing patterns of the distribution of information offered by research platforms and social networks. All practitioners in the information marketplace--libraries, publishers, and vendors--will continue to use whatever means they can to extract value from a stagnant set of resources. Publisher and vendor consolidation will continue, and libraries will actively manage their portfolios to get the biggest return for their dollars. The 6% average price increase seen in 2016 is expected to remain fairly constant for 2017, with perhaps a very slight downward trend hovering in the 5.5% to 6.5% range. Six percent seems to be a level of inflation that is neither too much for libraries to bear nor too little for publishers to accept.

The intent of this article is to present the most reliable information concerning serials pricing as is reasonably possible to produce. A March post on the Scholarly Kitchen blog, "Revisiting: Have Journal Prices Really Increased Much in the Digital Age?," which states the idea that serial prices have increased vastly more than the Consumer Price Index (CPI), is one of the most vexing pieces of misinformation about journal publishing. The vexing part is that the editors of the Scholarly Kitchen would use a study (Paula Gantz's "Digital Licenses Replace Print Prices as Accurate Reflection of Real Journals Costs") shown to be flawed to attempt to advance the idea that serial prices are not increasing faster than the CPI and that price studies are invalid.

Based on methodology established in the ANSI/NISO Z39.201999 Criteria for Price Indexes for Printed Library Materials, this price survey uses the available retail print or online price for subscriptions for titles that are listed in serials indexes widely used in the United States. The actual change in the retail price is measured year to year to gauge the overall increase in serial prices. This article includes tables and information documenting price changes for online journals as well as packages. All price increases for print, online, and packages are consistently in the 5% to 6% range annually.

The Gantz article referenced by the Scholarly Kitchen suggested serials price has now been replaced by digital licenses as the true reflection of real journal costs. In the article, the author directly questions the validity of this serial price study by suggesting that the "effective" price increase for an average journal was only 9% higher in 2010 as compared to 1990, not the six-fold increase documented in U's survey over the years.

Gantz raised good points concerning the increased value derived from digital licenses and how the increase in research has resulted in increased content. There does need to be more substantive discussion concerning these issues and the failure of libraries to secure their piece of the research funding pie. However, the contention that the effective price of the average journal has only increased 9% since 1990 and that the price of journals accessed in the UK has actually gone down 11% since 2004 were not substantiated in the article. After searching the reports cited in the UK study, no reference to the cited data could be found.

Anyone who has been involved with purchasing serials in the last 20 years knows that serial prices have increased significantly and represent the largest inflationary factor for library budgets. To suggest that the real increase in serial prices is only 9% flies in the face of objective reality.

The author derived the figure by using data from the Association of Research Libraries (ARL) that reported changes in the number of purchased serials and serial expenditures for the period 1990-2010. While the data does show that the numerical value resulting from dividing expenditures by the number of purchased serials did increase 9% over that period, to conclude that prices only increased 9% is incorrect. Between 1990 and 2010, ARL changed the definition of purchased serials. In 1990, the definition was based on subscribed titles. In 2010, the definition used by ARL had changed considerably: "[T]he total number of unique serial titles, NOT SUBSCRIPTIONS, that you currently acquire and to which you provide access.... Report each title once, regardless of how many subscriptions or means of access you provide for that title." (The definitions for 2010 and 2011 are basically the same.)

The majority of the increase in the number of serials purchased reported over this period is owing to the inclusion of titles purchased through databases and similar aggregations, not through "big deal" packages or other subscriptions. The cost of the access provided to these journals by libraries cannot be compared to the retail price of subscribed content.

The Value of Journals

Journal price data is important for budget management processes, but price alone is not the sole factor determining value. Some metrics such as Impact Factor have become important in assessing value, and similar metrics will only increase in importance in the future. Improvements in use data gathering (Counter 4) and developments in the altmetrics arena may provide even more data to help make qualitative assessments of the impact of scholarly publishing. Data-based decisions will be very important in determining values as libraries actively manage their information resources.

A project that will standardize altmetrics data continues to make progress as NISO released a draft of a proposed Altmetrics Data Quality Code of Conduct in 2016. A clear set of standards need to be in place to enable altmetrics to become a mainstream assessment tool for scholarly work, and this NISO effort will move the industry in that direction.

Until altmetrics matures, other standardized metrics will have to suffice.

This year we continue to work with the title and publisher data collected for this article to explore the relationship between prices and metrics used to assess journals, such as Impact Factor, Eigenfactor, and the Article Influence Score. This year, the relationship between serial prices and numbers of citations was explored with interesting results.

The Impact Factor of a journal is the average number of citations received per paper published in that journal during the two preceding years. The Impact Factor of a journal is calculated by dividing the number of current-year citations by the number of papers published in that journal during the previous two years.

The Eigenfactor rates journals according to the number of incoming citations, with citations from highly ranked journals weighted to make a larger contribution to the score than citations from poorly ranked journals. Journals are considered to be influential if they are cited often by other influential journals.

The Article Influence Score is determined by the average influence of a journal's articles over the first five years after publication. It is calculated by dividing a journal's Eigen-factor Score by the number of articles in the journal, normalized as a fraction of all articles in all publications. The mean Article Influence Score is 1.00. A score greater than 1.00 indicates that each article in the journal has above-average influence. A score less than 1.00 indicates that each article in the journal has below-average influence.

The pricing data in the merged ISI indexes for 2015 was divided into five price bands: journals priced $200 or less, journals priced between $201 and $550, $551 and $1,075,

$1,076 and $2,290, and titles priced at more than $2,290. These bands were selected only to ensure that the number of titles in each area was reasonably comparable. The average for Impact Factor, Eigenfactor, and Article Influence Score for all titles in a price range was compared to the averages in the other price bands. The results are displayed in Table 9 and the chart based on Table 9. The Impact Factor and the Eigenfactor tended to show a fairly strong rise with the increase in price. The Article Influence Score did not show a significant increase, with the average for titles less than $1,076 showing an average below the mean of 1 (below-average influence) and the more expensive titles showing an average of 1.2 (above-average influence).

Although there were increases in the metrics for Impact Factor and Eigenfactor, the increases were not comparable to the increase in price, since the average price ($5,055 for the most expensive journals) was 47 times higher than the least expensive ($107) journals. The increase in prices for the lower-cost titles was also lower than the more expensive titles. Higher-priced titles do have higher Impact Factors and Eigenfactors, but the increase in the metrics is small when compared to the increase in costs. Article Influence Score did not show a strong correlation between higher scores and prices. If the cost per citation is reviewed--the total cost of journals in a price band divided by total number of citations--high-priced titles had the highest cost per citation.

The ratio of citations to serial costs by subject is reviewed in Table 10. For STM journals, the average prices tend to be high in comparison to other subjects. This scenario changes if the costs are divided by the numbers of citations for the journals. Chemistry has the highest average price for journals, but it has the third lowest cost per citation. Journals in chemistry are very expensive, but they are heavily cited. If citations are considered an indicator of value, then chemistry journals, despite high average prices, are extremely high-value journals. Conversely, journals in the areas of philosophy, music, and literature are relatively cheap but are infrequently cited, so journals in those areas show the highest cost per citation.

Based upon this set of data, if cost per citation is reviewed by type of publisher, it is not surprising that commercial publishers have higher per citation costs than other types of publishers. Commercial publishers showed a cost per citation of 44 [cents] (average price $2,129), while university presses showed 200 per citation (average price $266), and societal publishers showed 11 [cents] (average price $1,377).

Stephen Bosch is Materials Budget, Procurement, and Licensing Librarian, University of Arizona Library, Tucson. Kittie Henderson is the Vice President for Academic, Law, and Public Library Markets, EBSCO Information Services, Birmingham, AL

TABLE 1 AVERAGE 2016 PRICE FOR SCIENTIFIC DISCIPLINES AVG. PRICE DISCIPLINE PER TITLE Chemistry $5,105 Physics 4,508 Engineering 3.244 Biology 3,104 Food Science 2,729 Astronomy 2,718 Botany 2,418 Geology 2,400 AVG. PRICE DISCIPLINE PER TITLE Technology $2,239 Zoology 2,221 Math & Computer Science 1,895 Health Sciences 1,801 General Science 1,717 Geography 1,713 Agriculture 1,687 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 TABLE 2 AVERAGE 2016 PRICE PER TITLE BY COUNTRY NO. OF AVG. PRICE COUNTRY ISI TITLES PER TITLE Russia 48 $5,103 Ireland 39 4,124 Hungary 13 3,641 Austria 23 3,552 Netherlands 572 3,093 Singapore 22 2,776 Switzerland 84 2,299 Germany 408 2,276 England 2,141 2,052 New Zealand 24 1,493 Greece 5 $1,283 United States 2,726 1,279 China 9 946 Australia 59 737 Norway 17 579 Sweden 13 558 Japan 50 528 Korea (South) 10 517 Canada 82 490 India 11 400 AVERAGE COST OF ALL TITLES: $1,818 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 TABLE 3 AVERAGE 2016 PRICE FOR ONLINE JOURNALS IN THE ISI INDEXES * AVG. PRICE DISCIPLINE PER TITLE Chemistry $4,712 Physics 4,180 Biology 2,704 Astronomy 2,665 Engineering 2,431 Botany 2,278 Zoology 2,250 Health Sciences 1,735 Geology 1,720 Technology 1,620 Math & Computer Science 1,618 Food Science 1,569 General Science 1,550 Agriculture 1,268 Geography 1,251 Military & Naval Science 1,002 Business & Economics 944 Psychology $942 Social Sciences 929 Education 897 Sociology 851 Political Science 776 Recreation 732 Law 580 General Works 551 Library Science 542 Anthropology 512 Arts & Architecture 510 Philosophy & Religion 484 History 466 Language & Literature 398 Music 334 TOTAL AVERAGE COST $1,486 * Prices represent print-plus-free-online, online-only, and the first tier of tiered pricing SOURCE: LJ PERIODICALS PRICE SURVEY 2016 TABLE 4 COMPARISON OF AVERAGE PRICE OF TITLES IN ISI INDEXES BY PRICE TO IMPACT FACTOR AVG. AVG. % PRICE OF LATEST NO. OF PRICE INCREASE IMPACT PRICE BAND TITLES 2016 2015-16 FACTOR Less than $201 919 $107 1.0 1.27 Between $201 and 550 1,429 375 4.6 1.72 Between $551 and 1,075 1,499 781 5.6 1.73 Between $1,076 and 2,290 1,600 1,564 6.2 2.60 Greater than $2,290 1,580 5,055 5.2 3.26 AVG. AVG. AVG. ARTICLE COST EIGEN- INFLUENCE PER PRICE BAND FACTOR SCORE CITATION Less than $201 0.0022 0.64 $0.22 Between $201 and 550 0.0040 1.00 0.24 Between $551 and 1,075 0.0057 0.85 0.29 Between $1,076 and 2,290 0.0135 1.09 0.25 Greater than $2,290 0.0237 1.22 0.40 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 TABLE 5 COMPARISON OF SERIAL PRICES WITH RATES OF CITATION BY LC SUBJECT NO. OF TOTAL TOTAL COST SUBJECT TITLES COST CITATIONS PER CITE Philosophy & Religion 220 $95,529 11,730 $8.15 Music 58 16,986 4,201 4.04 Language & Literature 460 176,021 64,442 2.73 Arts & Architecture 104 45,005 17,566 2.56 General Works 71 21,958 10,173 2.16 History 332 145,718 70,089 2.08 Library Science 55 47,382 35,095 1.35 Social Sciences 70 65,199 51,012 1.28 Military & Naval Science 13 13,831 11,824 1.17 Political Science 106 86,886 89,072 0.98 Law 115 55,058 57,178 0.96 Education 174 171,455 179,162 0.96 Recreation 48 35,140 37,417 0.94 Business & Economics 607 891,724 1,235,826 0.72 Sociology 344 351,918 507,238 0.69 Math & Computer Science 223 414,198 628,923 0.66 Anthropology 51 28,052 42,887 0.65 Geography 105 178,752 319,663 0.56 Zoology 120 262,527 478,197 0.55 Physics 227 1,021,014 2,118,148 0.48 Food Science 23 62,778 144,417 0.43 Botany 58 140,234 390,633 0.36 Astronomy 25 67,943 191,032 0.36 Engineering 438 1,399,378 4,103,485 0.34 Technology 89 181,777 541,173 0.34 Agriculture 174 288,109 881,072 0.33 Geology 94 223,244 732,660 0.30 Psychology 200 206,213 691,911 0.30 Biology 571 1,733,692 6,042,009 0.29 Chemistry 218 1,112,903 4,101,706 0.27 Health Sciences 1,528 2,594,689 11,329,039 0.23 General Science 102 161,893 1,497,450 0.11 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 Periodical Prices for High School and Public Libraries Overall price increases for titles in EBSCO Publishing's Masterfile Complete are expected to be in the 5% to 6% range TABLE 6 COST HISTORY FOR TITLES IN MASTERFILE COMPLETE AVG. AVG. % OF AVG. % OF NO. OF COST PER COST PER CHANGE COST PER CHANGE MASTERFILE TITLES TITLE TITLE 2014- TITLE 2015- COMPLETE 2014-16 2014 2015 15 2016 16 U.S. 1032 $252 $266 5.5 $279 4.8 NON-U.S. 234 339 359 6.0 376 4.7 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 Periodical Prices for University and College Libraries Overall price increases for titles in EBSCO's Academic Search Complete are expected to be in the 6% range for 2017 TABLE 7 2017 COST PROJECTIONS FOR TITLES IN ACADEMIC SEARCH COMPLETE ACADEMIC 2016 PROJECTED SEARCH NO. OF % OF AVG. COST % OF % OF COMPLETE TITLES LIST PER TITLE COST INCREASE U.S. 3,285 40 $966 37 5.8 NON-U.S. 4,890 60 1,615 63 5.6 ACADEMIC PROJECTED PROJECTED SEARCH 2017 AVG. % OF OVERALL % COMPLETE COST PER TITLE COST INCREASE U.S. $1,022 37 NON-U.S. 1,705 63 5.7 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 TABLE 8 TITLES INDEXED IN SCOPUS COST HISTORY BY LC SUBJECT AVG. AVG. AVG. NO. OF % OF COST COST TITLES CHANGE PER TITLE PER TITLE SUBJECT 2014-16 2014-16 2014 2015 Agriculture 456 -12 $952 $998 Anthropology 112 -2 506 535 Arts & Architecture 163 2 425 448 Astronomy 61 -1 1,746 1,863 Biology 1,110 -21 1,956 2,077 Botany 147 -3 1,170 1,238 Business & Economics 1,163 10 1,225 1,297 Chemistry 366 -4 3,667 3,830 Education 376 7 782 830 Engineering 1,698 16 1,586 1,661 Food Science 71 -3 1,512 1,639 General Science 208 -1 1,179 1,244 General Works 94 -1 331 336 Geography 266 5 952 991 Geology 207 -9 1,415 1,516 Health Sciences 3,915 -107 999 1,063 History 570 -9 357 437 Language & Literature 669 -6 350 367 Law 247 1 523 556 Library Science 126 1 921 965 Math & Computer Science 865 3 1,256 1,307 Military & Naval Science 61 0 566 624 Music 76 -1 272 287 Philosophy & Religion 377 8 360 375 Physics 407 -6 2,891 2,973 Political Science 212 4 636 668 Psychology 273 -11 709 769 Recreation 79 -2 578 621 Social Sciences 111 0 696 730 Sociology 636 -5 722 769 Technology 351 -1 1,384 1,454 Zoology 273 -8 1,160 1,268 AVERAGE $1,171 1,239 AVG. % OF COST % OF CHANGE PER TITLE CHANGE SUBJECT 2015 2016 2016 Agriculture 5 $1,069 7 Anthropology 6 567 6 Arts & Architecture 5 496 11 Astronomy 7 1,947 5 Biology 6 2,206 6 Botany 6 1,314 6 Business & Economics 6 1,374 6 Chemistry 4 4,025 5 Education 6 890 7 Engineering 5 1,788 8 Food Science 8 1,768 8 General Science 5 1,297 4 General Works 2 396 18 Geography 4 1,046 6 Geology 7 1,639 8 Health Sciences 6 1,132 7 History 22 460 5 Language & Literature 5 390 6 Law 6 560 1 Library Science 5 1,013 5 Math & Computer Science 4 1,359 4 Military & Naval Science 10 661 6 Music 6 304 6 Philosophy & Religion 4 395 5 Physics 3 3,127 5 Political Science 5 712 7 Psychology 8 834 9 Recreation 7 676 9 Social Sciences 5 820 12 Sociology 7 824 7 Technology 5 1,541 6 Zoology 9 1,366 8 AVERAGE 6 1,315 6 SOURCE: LJ PERIODICALS PRICE SURVEY 2016 TABLE 9 ISI INDEXES COST HISTORY BY LC SUBJECT AVG. AVG. AVG. NO. OF % OF COST COST TITLES CHANGE PER TITLE PER TITLE SUBJECT 2014-16 2014-16 2014 2015 Agriculture 170 0 $1,501 $1,598 Anthropology 50 -1 461 484 Arts & Architecture 104 -1 386 406 Astronomy 25 0 2,433 2,630 Biology 541 0 2,771 2,938 Botany 59 -1 2,159 2,264 Business & Economics 552 1 1,289 1,385 Chemistry 218 0 4,752 4,926 Education 169 0 855 911 Engineering 425 1 2,890 3,051 Food Science 23 0 2,382 2,546 General Science 79 1 1,504 1,638 General Works 70 0 248 252 Geography 101 0 1,506 1,609 Geology 92 1 2,122 2,263 Health Sciences 1,286 3 1,582 1,693 History 330 2 389 411 Language & Literature 458 2<