The opinion of the court was delivered by
Larson, J.:This is a tax protest appeal filed for the years 1998 and 2000 pursuant to K.S.A. 79-2005 involving eight tracts of pastureland in Rice County, Kansas, with valuations ranging from a high of $9,030 to a low of $1,680. The taxpayers, Ann W. Smith and others, challenge the methodology used by the Director of Property Valuation (PVD) of Kansas Department of Revenue and contend it is not in compliance with the statutory requirements of K.S.A. 79-1476.
The Board of Tax Appeals (BOTA) consolidated the appeals, joined the PVD as a necessary party, held á hearing, and denied the protests, holding the taxpayers had not presented sufficient evidence to refute die PVD’s agriculture land valuation methodology. After the taxpayers’ petition for reconsideration was denied they appealed to the Court of Appeals pursuant to K.S.A. 2000 Supp. 74-2426 and K.S.A. 77-601 et seq. Our jurisdiction is pursuant to K.S.A. 20-3018(c) (transfer from Court of Appeals on our own motion).
Contentions of the parties
The taxpayers narrowly contend (1) the nine crop reporting districts established by Kansas Agricultural Statistics (KAS) utilized by the PVD in its valuation formula are not homogeneous areas, (2) the use of soil qualities and types to determine gross rental income from pastureland does not comply with the K.S.A. 79-1476 requirement that value be based on net rental income normally received, and (3) the term “land classes” refers to usage and does not justify the use of soil types to classify and value land. These contentions lead to the broader claimed conclusion that the for*1398mula devised by the PVD does not comply with the mandates of K.S.A. 79-1476.
The PVD argues (1) the taxpayers did not meet their burden of proof to show the lands value differed from BOTA’s appraisal, (2) there was substantial competent evidence that PVD has established a proper classification system for all land devoted to agricultural use, using criteria established by the United States Department of Agriculture Soil Conservation Service and KAS, and (3) appraisal judgment and standards are followed and incorporated throughout the process of data collection and analysis in establishing pasture values pursuant to K.S.A. 79-1476.
Proceedings, evidence, and order of Board of Tax Appeals
The taxpayers began by protesting the manner of valuation of their pastureland for 1998 and 2000 as allowed by K.S.A. 79-2005. Their appeals and the evidence presented at their tax protest hearing contended that pasture rental is actually based on factors such as condition of the grass, carrying capacity, location, and fencing, but is never based on soil type. The taxpayers presented evidence that the area used by the PVD for their region runs from the Flint Hills grass on the east end to the short grass country on the west end and is not homogeneous as the statute requires. The taxpayers claimed their land should be appraised on actual rental rates in their county rather than by a complex formula that does not reflect rental rates for their property located in the center of the region utilized by the PVD. Their testimony failed to establish or opine what the actual valuation of their property should be.
The testimony of the PVD established that in determining the value of pasture or rangeland, the PVD combines information from several sources. KAS has divided the state into nine relatively equal sized crop reporting districts, with the divisions following county lines. The districts contain from 8 to 14 counties, and the district in question, C-50, has 11 counties. KAS surveys each district annually to find the average cash rental rate per acre of grassland. In 1997, the Department of Agricultural Economics of Kansas State University did a similar survey except that native and tame pastures were surveyed separately. From this survey, a ratio of native to *1399tame pasture cash rental rates was determined for each district. Using this ratio and the data gathered by KAS, an average cash rental for native and tame pasture per each district was calculated.
Dr. Paul Ohlenbusch of Kansas State University’s Department of Agronomy, in cooperation with the Natural Resource Conservation Service (NRCS) (which is the successor and synonymous to the Soil Conservation Service) developed a grazing index for each acre of pasture or rangeland based on soil type for both native and tame pasture.
The soil types are rated based on the forage producing capacity which determines the animal stocking rates. Bill Roth, a soil scientist, testified on behalf of the PVD that he helped to map soil types in the various counties. He noted that similar soil types are grouped together by the NRCS into range sites. He testified that the range sites were homogeneous regions. Mr. Roth stated that the forage rate is determined for each range site. Factors used to determine productivity of soil mapping units include “weather, soil characteristics (depth, texture, slope, fertility, moisture holding capacity), and plant growth and development.” Any further reference to soil classes, soil types, soil mapping units or range sites will be a reference to these groupings of soil types used to determine forage rates and the grazing index.
The PVD testimony and evidence included admission of a pamphlet prepared by Dr. Ohlenbusch which explains the process of developing the grazing index, and includes three basic assumptions:
“1. A single level of potential must be used for each type of grazing land.
“2. The level selected must be attainable by a reasonable individual with an average knowledge of the management requirements of the type of grazing land they use.
“3. The future of the grazing land’s potential must be a consideration in the process.”
To develop the index, two existing databases together with published soil surveys were used. The current listing of soil series by county was obtained from the NRCS and combined with a range-land carrying capacity database developed by Kansas State University Extension and NRCS personnel in the mid-1980’s. The carrying capacity database was based on NRCS range site descriptions *1400which are a part of the standard materials maintained in field offices. Values are assigned by clipping existing vegetation on representative range sites in different conditions to determine the dry matter produced. The soil surveys provided the basis for the tame pasture data. In addition to the following basic requirements, the index assumes that average grazing management is used.
The classification system is based on plant succession which states that plant species change in response to external factors. The major natural factors are drought, grazing, and fire. Today, management is the dominant factor. The rangeland was indexed at a good condition class which, based on research and experience, is considered to be the most profitable condition level to manage that can also maintain and potentially improve the resource.
Average soils in a district carry an index value of one, while the more productive soils are rated higher and vice versa. The average gross rental income per acre for each district is determined by surveying the actual rates paid by individuals within the districts. To find the gross rental income of a particular acre of ground, the weighted average grazing rate of the parcel is multiplied by the average gross rental income of the district. Average expenses including maintenance, management, water, and fencing costs, determined on a per acre basis for each district, are deducted from the gross rental income to reach net rental income per soil mapping unit per district each year. The net rental income for the previous 8 years are averaged together, and this amount is submitted to the PVD.
The PVD capitalizes the 8-year average net rental income at a rate provided by K.S.A. 79-1476 to arrive at the “use-value.” The PVD determines the use-value (the taxable value) per acre of range site, per district, for both native and tame pasture and then provides those values to the county appraisers. The PVD representative testified that the values are “fine tuned” for each county based on range sites located in the counties and clippings taken from those sites. Values are also calculated for dry land and irrigated farm acreage, neither of which are the subject of this appeal. The county appraiser classifies the property as native pasture, tame pasture, cultivated dry land, or cultivated irrigated land (or some frac*1401tion of each), then multiplies the appropriate use-value provided by the PVD times the number of acres of each soil type on the parcel to ascertain the assessed value of the property.
Zoe Gehr, the PVD representative, testified die crop reporting districts were homogeneous in a geographical sense. Mr. Roth testified the range sites used for soil classification were homogeneous areas. All the witnesses agreed the crop reporting districts were not homogeneous as to soil quality and productivity.
After hearing this testimony, and reviewing the exhibits submitted, BOTA reached the following findings and order:
“5. The subject parcels consist of eight (8) agricultural parcels located in Rice County, Kansas with respective agricultural use value as listed above in paragraph 2. The Taxpayers have protested the State’s methodology utilized in valuation of these parcels for the 1998 and 2000 tax years. The Taxpayers questioned numerous elements of PVD’s methodology utilized to value the subject agricultural use parcels.
“6. Don Kueck and Wayne Smith appeared as witnesses for the Taxpayers and testified as follows: Soil types have some effect on rental rates, yet soil type alone is not the sole determinant for land livestock carrying capacity. Similar soil types in different areas may have different productivity due to the amount of rain, temperature, hillside exposure, and/or wind/water erosion. Poorer soil types often produce better grasses for livestock. A land manager must stock livestock based on land productivity potential with consideration for the more fragile soils. For a landlord, there are circumstances where a parcel’s shape results in a larger section being more economically efficient than a smaller section due to the fencing costs. Also, certain communities have different stocking rates based primarily on demand, location, and proximity to water supply rather than on land soil types. Lastly, the Taxpayers argued that the central crop reporting district is not homogeneous and should be separated into three (3) difference [sic] regions. See Taxpayer Exhibit No. 1.
“7. Truett McQueen appeared on behalf of Rice County and testified as follows: The subject parcels were valued in accordance with PVD guidelines. The County informally performed an agricultural survey and determined that actual stocking rates were lower than that indicated by PVD. See Taxpayer Exhibit No. 2. Mr. McQueen submitted that he could not adjust the agricultural use values implemented by PVD. Mr. McQueen submitted a disparity in agricultural use values in Rice County. Mr. McQueen also submitted that cultivated land provides a higher economic return than pastureland, yet PVD has certain pastureland that is valued greater than certain types of cultivated land. In regard to the Taxpayers’ parcels, Mr. McQueen testified *1402that while the subject parcels have a mix of soil types, different rental rates are not charged based on said sod types. Any difference in pasture rates is due to historical rates and the availability of water.
“8. PVD testified and submitted evidence to indicate as follows: The 1998 and 2000 agricultural use values for the subject properties were determined in compliance with K.S.A. 79-1476. Rice County is located in crop reporting district C-50, which is composed of 11 counties. Kansas Agriculture Statistics and Kansas State University survey the crop reporting districts to obtain a statistically significant number of responses, and this data is utilized in valuing agricultural use land. PVD further explained how pasture and rangeland agricultural use values are calculated. See PVD Exhibits No. 1 through 3.
“9. Land devoted to agricultural use is valued pursuant to Kan. Const, art. XI, §12 and K.S.A. 79-1476, and amendments thereto. The valuation method is based upon the income or agricultural production capabilities. Types of soil are to be classified into homogeneous groups so that the values will be uniform and equal. The Kansas Director of the Division of Property Valuation (PVD) is to calculate the eight-year average production immediately preceding the calendar year which immediately precedes the year of valuation. See K.S.A. 79-1476, and amendments thereto. Reasonable agricultural expenses should be subtracted from the landlord’s share of the net rental income. The net income is then capitalized at a rate determined to be the sum of the contract rate of interest on new federal land bank loans in Kansas on July 1 of each year averaged over a five-year period. . . . ’ See K.S.A. 79-1476, and amendments thereto.
“10. The powers of the director of Property Valuation are addressed in several statutes. The director has general supervision of the taxation system throughout the state. See K.S.A. 79-1402. Pursuant to K.S.A. 79-1404 Sixteenth, the director is compelled to:
[d]o and perform any act or to make any order or direction to any county board of equalization or any county or district appraiser as to the valuation of any property or any class of property in any township, city or county which, in die judgment of said director of property valuation, may seem just and necessary, to the end diat all property shall be valued and assessed in the same manner and to the same extent as any and all other property, real or personal, required to be listed for taxation.
“11. K.S.A. 79-1476 empowers the director to ‘administer and supervise a statewide program of reappraisal of all real property located within the state.’
“12. K.S.A. 79-1456 requires county appraisers to ‘follow the policies, procedures and guidelines of the director of property valuation in the performance of the duties of the office of county appraiser.’
“13. The Board finds that pursuant to K.S.A. 79-1402, K.S.A. 79-1404 and K.S.A. 79-1476, the director is empowered by the legislature to use the Kansas Agricultural Statistics’ Crop Reporting Districts and Division of Water Resources Irrigation Districts. Pursuant to this same grant of power, the Board *1403finds that the director has the authority to use the methodologies as illustrated in PVD Exhibit No. 1.
“14. The Board further finds that PVD has complied with K.S.A. 79-1476 in valuing agricultural land, and that the values determined by PVD are the best estimate of use value for the subject parcels.
“15. Pursuant to K.S.A. 79-1609, the County Appraiser must support the validity and correctness of the value by a preponderance of evidence for residential property or real property used for commercial and industrial purposes. Pursuant to Kan. Const, art. XI, § 1, the subject property is not classified as residential property or real property used for commercial and industrial purposes. Therefore, the burden of demonstrating the incorrectness of the value is on the Taxpayer.
“16. The Board finds that the Taxpayers have presented general information and referenced various instances concerning elements of PVD’s valuation methodology; however, the Taxpayers have not presented any evidence which would indicate that the subject parcels have not been valued in accordance with applicable State statutes. Additionally, the Board notes that upon cross-examination, Mr. McQueen acknowledged the imprecision of the County’s agricultural survey. Summarily, the Board finds that the Taxpayers have not presented sufficient evidence to refute PVD’s agricultural land valuation methodology. The Board further finds that the evidence presented by PVD substantiates the subject parcels’ present respective 1998 and 2000 agricultural use values. Based thereon, the Board concludes that the subject parcels’ present respective 1998 and 2000 agricultural use values shall be, and the same are hereby, sustained.
“IT IS THEREFORE ORDERED BY THE BOARD OF TAX APPEALS OF THE STATE OF KANSAS that, for the reasons stated above, the Taxpayers’ protests shall be, and the same are hereby, denied.”
Standard of review
The taxpayers’ challenge relates to the interpretation and application of K.S.A. 79-1476. That is a question of law over which this court exercises plenary review. Babe Houser Motor Co. v. Tetreault, 270 Kan. 502, 506, 14 P.3d 1149 (2000).
In interpreting the applicable statutes, we look to the rules set forth in Todd v. Kelly, 251 Kan. 512, 515-16, 837 P.2d 381 (1992):
“ ‘The function of the court is to interpret the statutes, giving the statutes the effect intended by the legislature. State ex rel. Stephan v. Kansas Racing Comm’n, 246 Kan. 708, 719, 792 P.2d 971 (1990).
“ ‘As a general rule, statutes are construed to avoid unreasonable results. Wells v. Anderson, 8 Kan. App. 2d 431, 659 P.2d 833, rev. denied 233 Kan. 1093 (1983). There is a presumption that the legislature does not intend to enact useless or *1404meaningless legislation. In re Adoption of Baby Boy L., 231 Kan. 199, Syl. ¶ 7, 643 P.2d 168 (1982).’ City of Olathe v. Board of Zoning Appeals, 10 Kan. App. 2d 218, 221, 696 P.2d 409 (1985).
“ ‘A construction of a statute should be avoided which would render the application of a statute impracticable or, inconvenient or, which would require the performance of ... an impossible act.’ In re Adoption of Baby Boy L., 231 Kan. 199, Syl. ¶ 8, 643 P.2d 168 (1982). See 73 Am. Jur. 2d, Statutes § 251.
“ ‘In construing statutes, the legislative intention is to be determined from a general consideration of the entire act. Effect must be given, if possible, to the entire act and every part thereof. To this end, it is the duty of the court, as far as practicable, to reconcile the different provisions so as to make them consistent, harmonious, and sensible.” In re Marriage of Ross, 245 Kan. 591, 594, 783 P.2d 331 (1989).
“ ‘[T]he court must give effect to the legislature’s intent even though words, phrases or clauses at some place in the statute must be omitted or inserted.’ Ross, 245 Kan. at 594.
“ ‘In order to ascertain the legislative intent, courts are not permitted to consider only a certain isolated part or parts of an act, but are required to consider and construe together all parts thereof in pari materia. . . .’ Kansas Commission on Civil Rights v. Howard, 218 Kan. 248, Syl. ¶ 2, 544 P.2d 791 (1975).”
The State contends BOTA’s ruling that the taxpayers have not presented sufficient evidence to refute PVD’s agricultural land valuation methodology is a negative ruling. We have held that when a trial court makes a negative factual finding the party challenging that finding must prove arbitrary disregard of undisputed evidence or some extrinsic consideration such as bias, passion, or prejudice because the negative finding signifies the failure of the party upon whom the burden of proof was cast to sustain it. Thomason v. Stout, 267 Kan. 234, 238, 978 P.2d 918 (1999).
The review we make of decisions of BOTA in valuation matters was set forth in Board of Ness County Comm’rs v. Bankoff Oil Co., 265 Kan. 525, 537, 960 P.2d 1279 (1998):
“Our standard of review of the actions of BOTA is pursuant to K.S.A. 77-621(c). In re Tax Appeal of Boeing Co., 261 Kan. 508, 514-15, 930 P.2d 1366 (1997). We may only grant relief if we find:
‘(1) The agency action, or the statute or rule and regulation on which the agency action is based, is unconstitutional on its face or as applied;
‘(2) the agency has acted beyond the jurisdiction conferred by any provision of law;
*1405‘(4) the agency has erroneously interpreted or applied the law;
‘(5) the agency has engaged in an unlawful procedure or has failed to follow prescribed procedure;
‘(7) the agency action is based on a determination of fact, made or implied by the agency, that is not supported by evidence that is substantial when viewed in light of the record as a whole, which included the agency record for judicial review, supplemented by any additional evidence received by the court under this act; or
‘(8) the agency action is otherwise unreasonable, arbitrary or capricious.’ K.S.A. 77-621(c).
“In Boeing, we said: ‘BOTA is a specialized agency that exists to decide taxation issues. BOTA’s decisions should be given great deference when it is acting in its area of expertise. However, if we find that BOTA’s interpretation is erroneous as a matter of law, we should take corrective steps.’ 261 Kan. 508, Syl. ¶ 3. A party challenging the validity of BOTA’s action has the burden of proving it was erroneous. In re Tax Appeal of Scholastic Book Clubs, Inc., 260 Kan. 528, 536, 920 P.2d 947 (1996).”
K.S.A. 79-1476
The statute in issue is long and complicated and is set forth in applicable part as follows:
“The director of property valuation is hereby directed and empowered to administer and supervise a statewide program of reappraisal of all real property located within the state. Except as otherwise authorized by K.S.A. 19-428, and amendments thereto, each county shall comprise a separate appraisal district under such program, and the county appraiser shall have the duty of reappraising all of the real property in the county pursuant to guidelines and timetables prescribed by the director of property valuation and of updating the same on an annual basis. In the case of multi-county appraisal districts, the district appraiser shall have the duty of reappraising all of the real property in each of the counties comprising the district pursuant to such guidelines and timetables and of updating the same on an annual basis. Commencing in 1994, every parcel of real property shall be actually viewed and inspected by the county or district appraiser once every four years. Any county or district appraiser shall be deemed to be in compliance with the foregoing requirement in any year if 25% or more of the parcels in such county or district are actually viewed and inspected.
“Valuations shall be established for each parcel of real property at its fair market value in money in accordance with the provisions of K.S.A. 79-503a, and amendments thereto.
“In addition thereto valuations shall be established for each parcel of land devoted to agricultural use upon the basis of the agricultural income or productivity attributable to the inherent capabilities of such land in its current usage under a *1406degree of management reflecting median production levels in the manner hereinafter provided. A classification system for all land devoted to agricultural use shall be adopted by the director of property valuation using criteria established by the United States department of agriculture soil conservation service. For all taxable years commencing after December 31, 1989, ah land devoted to agricultural use which is subject to the federal conservation reserve program shall be classified as cultivated dry land for the purpose of valuation for property tax purposes pursuant to this section. Productivity of land devoted to agricultural use shall be determined for all land classes within each county or homogeneous region based on an average of the eight calendar years immediately preceding the calendar year which immediately precedes the year of valuation, at a degree of management reflecting median production levels. The director of property valuation shall determine median production levels based on information available from state and federal crop and livestock reporting services, the soil conservation service, and any other sources of data that the director considers appropriate.
“The share of net income from land in the various land classes within each county or homogeneous region which is normally received by the landlord shall be used as the basis for determining agricultural income for all land devoted to agricultural use except pasture or rangeland. The net income normally received by the landlord from such land shall be determined by deducting expenses normally incurred by the landlord from the share of the gross income normally received by the landlord. The net rental income normally received by the landlord from pasture or rangeland within each county or homogeneous region shall be used as the basis for determining agricultural income from such land. The net rental income from pasture and rangeland which is normally received by the landlord shall be determined by deducting expenses normally incurred from the gross income normally received by the landlord. Commodity prices, crop yields and pasture and rangeland rental rates and expenses shall be based on an average of the eight calendar years immediately preceding the calendar year which immediately precedes the year of valuation. Net income for every land class within each county or homogeneous region shall be capitalized at a rate determined to be the sum of the contract rate of interest on new federal land bank loans in Kansas on July 1 of each year averaged over a five-year period which includes the five years immediately preceding the calendar year which immediately precedes the year of valuation, plus a percentage not less than .75% nor more than 2.75%, as determined by the director of property valuation.
“Based on the foregoing procedures the director of property valuation shall make an annual determination of the value of land within each of the various classes of land devoted to agricultural use within each county or homogeneous region and furnish the same to the several county appraisers who shall classify such land according to its current usage and apply the value applicable to such class of land according to the valuation schedules prepared and adopted by the director of property valuation under the provisions of this section.
*1407“It is the intent of the legislature that appraisal judgment and appraisal standards be followed and incorporated throughout the process of data collection and analysis and establishment of values pursuant to this section.
“The provisions of this act shall not be construed to conflict with any other provisions of law relating to the appraisal of tangible property for taxation purposes including the equalization processes of the county and state board of tax appeals.”
Legislative overview of K.S.A. 79-1476 and attempted amendments
In 1975, article 11, § 12 was added to the Kansas Constitution. L. 1975, ch. 516, § 1. It permits land devoted to agricultural use to be valued “for ad valorem tax purposes upon the basis of its agricultural income or agricultural productivity, actual or potential.” Prior to that time, fair market value was used. In 1985, the legislature, pursuant to the authority granted in Kan. Const, art. 11, § 12 enacted K.S.A. 79-1476. L. 1985, ch. 314, § 1; see also House Committee of Assessment and Taxation minutes, March 22, 1985, attachment 10. A representative of the Kansas Livestock Association testified on behalf of Senate Bill 164. He stated that the proposed statute might raise the rates of some agricultural land, but the bill was still favored because “land would be appraised on a more mathematically certain basis and more importantly, on a more logical and equitable basis that has some relationship to the income that farmers and ranchers are capable of earning from it.” House Committee of Assessment and Taxation minutes, March 22, 1985, attachment 10. He also stated that the purpose was to prevent farmers and ranchers from being forced to sell based on excessive real estate taxes due to rising prices of neighboring urban properties.
A representative from the Kansas Department of Revenue real estate division presented an analysis of agricultural land “use value” to the House Assessment and Taxation Committee. He noted that the Soil Conservation Service identified eight “land capability classes.” A sample of 13 counties was analyzed to determine use value, per county, per land capability class. Fair market values were also determined for each class. Only two counties were analyzed to determine use value and fair market value for the irrigated land *1408classes. He defined “use value” and how that is determined for rangeland:
“ ‘Use Value’ can have more than one meaning; however, it is generally conceded, as it relates to agricultural land, that ‘Use Value’ represents a value based on the net income that the land is capable of producing, assuming typical management practices.
“The procedures used to estimate the net income for the various land capability classes are as follows:
“11. Determine the gross cash rent paid per acre for the various rangeland classes. The gross cash rent is an average of that amount paid over the past eight years.
“12. Estimate the typical landlord expenses for each of the rangeland classes. This amount is an average of those expenses incurred over the past eight years.
“13. Deduct the expenses from the gross income for each of the various range-land classes and process die net income into capital value.”
Minutes of House Committee on Assessment and Taxation, March 20, 1985, attachment 1.
A supplemental note to the Senate Bill 164, also introduced during discussion of the bill by the House Assessment and Taxation Committee on March 19,1985, summarizes the effect of the agricultural provisions: “Values for land devoted to agricultural use also would be established on the basis of agricultural productivity (use-value).”
In 1990, the legislature amended K.S.A. 79-1476, adding the following language: “For all taxable years commencing after December 31, 1989, all land devoted to agricultural use which is subject to the federal conservation reserve program shall be classified as cultivated dry land for the purpose of valuation for property tax purposes pursuant to this section.” (Emphasis added.) L. 1990, ch. 347, § 1. In 2000, the legislature added a similar provision: “For all taxable years commencing after December 31, 1999, all land devoted to agricultural use which is subject to the federal wetlands reserve program shall be classified as native grassland for the purpose of valuation for property tax purposes pursuant to this section.” (Emphasis added.) L. 2000, ch. 139, § 7.
Also in 2000, the legislature rejected an attempted amendment to K.S.A. 79-1476 which would have expressly prohibited use of soil classifications for determining net income from pasture and *1409rangeland. The amendment also would have expressly permitted the county or district appraiser to consider adverse factors in valuing individual parcels of ground:
“The net rental income normally received by the landlord from pasture or range-land within each county or homogeneous region shall be used solely as the basis for determining agricultural income from such land, and soil classifications for such land shall not be considered. . . .
“ . . . Notwithstanding the foregoing, any county or district appraiser may apply adverse influence factors to any such value and deviate from such value accordingly subject to [review and approval] of the assistant director of property valuation.” H. B. 2715 (2000).
The bill was passed by the House but died in the Senate Committee on Taxation and Assessment.
Attorney General Opinion No. 89-63
In 1989, former Representative Gayle Mollenkamp requested an opinion from the Attorney General as to whether the PVD’s use of quality of the soil in valuing pastureland was proper. He contended the PVD’s actions were in violation of K.S.A. 1988 Supp. 79-1476, which requires “net rental income normally received” to the basis of valuation. He stated: “Pastureland is pastureland, regardless of quality of location.”
In response, the Attorney General issued opinion No. 89-63 which in its analysis of K.S.A. 1988 Supp. 79-1476 pointed out that the statute requires the PVD “to develop a classification system for all land devoted to agricultural use,” and that “the statute does not except pasture and rangeland from this provision.” The opinion further states that the criteria for the classification system by statute must come from the Soil Conservation Service (SCS), and the criteria that agency has developed is based on soil types:
“The land is to be classified using criteria established by the . . . SCS. SCS criteria consists of soil types with the same or similar production capability assigned to a specific productivity group. Thus, in a sense, the ‘quality’ of the soil is the basis used for determining classification under the statute.” Att’y Gen. Op. No. 89-63.
The Attorney General approved the PVD’s classification of pasture and rangeland by soil quality and the use of that system to *1410value grassland: “In our opinion, the procedure developed by the [PVD] for valuing pasture and rangeland ... is within the statutory guidelines. It appears to adequately derive median values for various types of land.” (Emphasis added.) Att’y Gen. Op. No. 89-63.
Analysis
Although a question of jurisdiction was raised, both parties now contend and we agree that jurisdiction does exist in the appellate courts. Taxpayers are appealing an appraisal and while the role of the county appraiser and the PVD might not be as clear as the parties believe, the assessment appears to be made by the PVD. Pursuant to K.S.A. 2000 Supp.74-2426, jurisdiction for the appeal is in the Court of Appeals from which our transfer was made.
This then brings into focus the contentions of the taxpayers that (1) the nine crop reporting districts are not compliant with the statutoiy requirement that net rental income be drawn from each “county or homogeneous region,” and more specifically that the districts are not “homogeneous,” (2) the use of soil qualities to determine the gross rental income for pasture and rangeland is not compliant with the statute’s requirement that value be based on “net rental income normally received,” and (3) the term “land classes” in K.S.A. 79-1476 refers to cultivated dry or irrigated land and native and tame pasture, not soil types, and the language does not justify the use of soil types to classify and value land.
K.S.A. 79-1476 requires that
“valuations . . . be established for each parcel of land devoted to agricultural use upon the basis of the agricultural income or productivity attributable to the inherent capabilities of such land in its current usage under a degree of management reflecting median production levels in the manner hereinafter provided. A classification system . . . shall be adopted by the director of property valuation using criteria established by the United States department of agriculture soil conservation service.” (Emphasis added.)
As we earlier outlined, the system utilized by the PVD determines the productivity of land at a median or lesser degree of management. This system classifies land by soil type, and produc*1411tivity levels are calculated for potential use of the land. This system generally appears to comply with the statute.
However, K.S.A. 79-1476 provides more specific guidance, as it later provides:
"Productivity of land devoted to agricultural use shall be determined/or all land classes within each county or homogeneous region based on an average of the eight calendar years immediately preceding the calendar year which immediately precedes the year of valuation, at a degree of management reflecting median production levels. The director of property valuation shall determine median production levels based on information available from state and federal crop and livestock reporting services, the soil conservation service, and any other sources of data that the director considers appropriate.” (Emphasis added.)
PVD’s present methodology utilizes the soil classifications from each crop reporting district. It contends this is a homogeneous region. Taxpayers contend these areas are too large, have too much variance between eastern and western boundaries (especially in the district which includes Rice County), and do not comply with the statutory requirements.
The areas selected are an attempt by the PVD to determine landlord net rental income by regions large enough to provide statistically accurate results and at the same time satisfy the longevity requirement. The regions selected are those utilized by KAS for purposes of its agricultural surveys. Within these regions the productivity of the soil types is determined and adjusted based on long-term clipping studies of the foliage produced by soil scientists who are range management specialists. From these environmental studies and clipping results, a grazing index is generated which approximates carrying capacity for each soil type throughout the state. When adjustment is made throughout the district based on the application of the grazing index, the mandates of the statute appear to be followed.
KAS also determines the landlord gross rental income by annual surveys that are randomly obtained throughout the district. The surveys appear to be statistically more justifiable than the survey of a single class of property in Rice County about which testimony was given as a part of the taxpayers’ protest. BOTA noted that *1412taxpayers’ witness McQueen acknowledged the imprecision of the county’s agricultural survey.
The testimony of PVD’s witnesses at the BOTA hearing relating to this issue was presented by Bill Roth, who believed die range sites used for soil classification were homogeneous areas, while Zoe Gehr opined the crop reporting districts were homogeneous in a geographical sense.
The BOTA order was not precise on the taxpayers’ “homogeneous” contention, as it stated generally in paragraph 16 that “[t]he Board finds that the taxpayers have presented general information and referenced various instances concerning elements of PVD’s valuation methodology, however, the taxpayers have not presented any evidence which would indicate that the subject parcels have not been valued in accordance with applicable State statutes.” We can only take this as a finding that the homogeneous element of the PVD methodology was approved by BOTA.
There is no definition of homogeneous in the statute, but our normal rules of construction require words to be given their ordinary meaning. Matjasich v. State Dept. of Human Resources, 271 Kan. 246, 252, 21 P.3d 985 (2001). “Homogeneous” is defined as “[o]f the same or similar nature or kind.” Webster’s II, New Riverside University Dictionary 589 (1984). Counties comprising crop reporting district C-50 have a similar geographical location, as each is located in Central Kansas, but in respect to productivity the district is not uniform, with the Flint Hills being on the eastern boundaries and short grass country found on the western boundary. The PVD’s testimony explains that these differences are made “homogeneous” by the usage of the grazing index and the adjustments it makes.
When we look at the legislative language we see that PVD may utilize counties or homogeneous regions upon which to base “net income.” The taxpayers’ own witness, Mr. Kueck, admitted that even counties are not completely homogeneous insofar as productivity and other factors could cause different rental rates of two pastures directly adjacent to each other.
We conclude there was substantial evidence to uphold BOTA’s findings that PVD was following statutory guidelines in utilizing *1413the districts into which KAS has divided the state of Kansas for purposes of gathering valid statistical information. To limit the mass appraisal system to only a county-by-county approach was not legislatively required and would be an extremely expensive proposition. We rely on BOTA’s expertise in examining PVD’s actions and finding it met the statutoiy requirements of K.S.A. 79-1476.
We will also consider the taxpayers’ contention that net income for a parcel of property must be derived from actual and not potential income of the agricultural land. It is clear that PVD’s formula does incorporate an actual average gross rental income in calculating use value, but there is a question as to whether K.S.A. 79-1476 is limited only to the utilization of actual rental figures. The taxpayers’ argument is based on that portion of K.S.A. 79-1476 which states:
“The net rental income normally received by the landlord from pasture or range-land within each county or homogeneous region shall be used as the basis for determining agricultural income from such land. The net rental income from, pasture and rangeland which is normally received by the landlord shall be determined by deducting expenses normally incurred from the gross income normally received by die landlord.” (Emphasis added.)
Taxpayers’ contention is similar to those raised by Representative Mollenkamp in his request for the Attorney General’s opinion which we have referred to previously. As we have previously stated, such an opinion is at times persuasive but is not binding on this court. City of Junction City v. Cadoret, 263 Kan. 164, 173, 946 P.2d 1356 (1997). In that opinion, the usage of the “quality” of the soil as a basis for determining classification was approved. The PVD procedure was approved, and the opinion stated that “it appears to adequately derive median value for various types of land.” Att’y Gen. Op. No. 89-63.
Net rental income normally received is to be used as the basis for valuing pasture and rangeland, and it is calculated by deducting expenses normally incurred from gross income normally received. K.S.A. 79-1476. If read in isolation, this portion of the statute could bring into doubt the PVD’s use of soil quality in addition to average actual gross income as a basis for valuing grassland. However, it is the duty of this court to construe several provisions of an Act to*1414gether, “with a view of reconciling and bringing them into workable harmony.” State v. Bolin, 266 Kan. 18, 24, 968 P.2d 1104 (1998).
K.S.A. 79-1476 requires that the PVD develop a classification system for all land from criteria established by the SCS and that valuations be established upon “the inherent capabilities of such land.” The statute also requires that net income “for every land class” be capitalized in calculating a taxable value. (Emphasis added.) Hence, the required calculation of gross income “normally received” by landlords in a county or homogeneous region can and must be reconciled with the statute’s earlier requirement that all agricultural land be classified. The PVD has made an effective attempt to accomplish both of these requirements by first classifying land by quality, then calculating the average actual gross income received by landlords per crop reporting district. The average actual gross income is multiplied by a weighted average system developed from the soil quality classification to reach the “normal” gross income.
Taxpayers’ request that actual gross rental income be used in determining “normal gross income” is actually neither less nor more precise that the manner PVD has developed. While we have previously stated the deference which is to be given to administrative agencies, we again set forth our obligation to allow operative construction in cases such as this. We recently said in McTaggert v. Liberty Mut. Ins., 267 Kan. 641, 645, 983 P.2d 853 (1999):
“ ‘The interpretation of a statute by an administrative agency charged with the responsibility of enforcing that statute is entitled to judicial deference. . . . Further, if there is a rational basis for the agency’s interpretation, it should be upheld on judicial review. If, however, the reviewing court finds that the administrative body’s interpretation is erroneous as a matter of law, the court should take corrective steps. The determination of an administrative body as to questions of law is not conclusive and, while persuasive, is not binding on the courts.’ [Citation omitted.]”
Once again, we emphasize that when called upon to develop a system of mass appraisal PVD was required to develop a system which unfortunately lost individualized treatment of the determination of use value. We agree with the 1989 opinion of the Attorney General. The PVD formula combines the sampling and polling *1415done by KAS of actual received rental income for property within the district polled with the grazing index from which a sufficiently accurate net rental income is obtained to be utilized in the process of determining the use value of pastureland.
We do not find there to be sufficient merit in the taxpayers’ argument that the entire PVD methodology must be struck down based on the statutory reference to land class. Taxpayers contend this relates only to native and tame pasture and dry and irrigated cultivated land. The PVD contends that land class has a dual meaning referring to both use classes as well as productivity classes.
K.S.A. 79-1476 states:
“A classification system for all land devoted to agricultural use shall be adopted by the director of property valuation using criteria established by the United States department of agriculture soil conservation service. For all taxable years commencing after December 31, 1989, all land devoted to agricultural use which is subject to the federal conservation reserve program shall be classified as cultivated diy land for the purpose of valuation for property tax purposes pursuant to this section.”
The first sentence envisions developing classes of land based on agricultural quality as the PVD has done by using soil types to classify and value similar areas of property. The second sentence justifies the taxpayers’ argument, as it speaks of classifying Conservation Reserve Program land as cultivated dry land.
Actually, the PVD uses productivity to classify land and also appears to utilize different classifications as to land usage. We do not believe either to be improper.
When originally enacted in 1985, K.S.A. 79-1476 did not include the following sentence: “For all taxable years commencing after December 31, 1989, all land devoted to agricultural use which is subject to the federal conservation reserve program shall be classified as cultivated dry land for the purpose of valuation for property tax purposes pursuant to this section.” See L. 1985, ch. 314, § 1. That language was not added until 5 years later. See L. 1990, ch. 347, § 1. Hence, “classifications” did not encompass use classes in the original 1985 enactment.
We hold that a reasonable interpretation of the statute supports a dual meaning of classification and land classes. It is illogical to *1416limit the classification system to land usage when the statute required the PVD to adopt a classification system based on soil composition as developed by the expertise of the Soil Conservation Service.
We do not give the credence requested by the PVD to the failure of the legislature to amend the specific provisions of 79-1476 as a total vindication of each and every portion of the methodology developed. This does, however, point out that mass appraisal and reappraisal are issues much better handled by legislative action and the expertise of administrative agencies than by the courts of our state.
If taxpayers desire to force the changes which they champion here, we believe the proper forum is legislative action as has apparently been attempted based on the failed amendment from 2000 which we previously discussed. Adoption by the court of the taxpayers’ arguments would totally upset the current appraisal procedures. Based on the limited and conflicting evidence presented, BOTA reached a result in its resolution of this taxpayer appeal that must be approved.
Finally, although we have not chosen to resolve this appeal on the failure of the taxpayers to provide any specific evidence of the actual values which they contend should have been applied to their property, such argument would likewise require that the BOTA decision be approved.
Affirmed.
Davis, J., not participating. Brazil, Chief Judge Retired, assigned.