[¶ 1] Lance Morris appeals from a district court judgment awarding Ceisha Mol-*268ler primary residential responsibility of the parties’ two children. We conclude the court’s judgment was not clearly erroneous, and we affirm.
I
[¶ 2] Morris and Moller were not married but lived together from the latter part of 2004 through February 2009. They had two children — C.M., a son, and J.M., a daughter — before separating. Moller also had a daughter, A.G., from a previous relationship; A.G. resided with the parties and their children. After the parties separated, Morris filed suit for primary residential responsibility of the children. Moller answered, seeking primary residential responsibility herself, and she moved for an interim order awarding her temporary primary residential responsibility. The court entered an interim order granting Moller’s motion and allowing Morris reasonable visitation. Following a motion by Moller, the district court ordered the appointment of a parenting investigator.
[¶ 3] On July 22 and 23, 2010, a trial was held to determine parental rights and responsibilities. Both parties alleged the other had abused drugs during their relationship. Morris asserted Moller used methamphetamine during her pregnancies with C.M., who was born in 2005, and J.M., who was born in 2007, which Moller denied. Among the witnesses called at trial were the children’s pediatrician and the parenting investigator, the latter of whom Morris claimed was biased toward Moller. After analyzing the best interests and welfare of the children under the factors set forth in N.D.C.C. § 14-09-06.2(1), the district court found it was in the children’s best interests for Moller to have primary residential responsibility and for Morris to have reasonable parenting time.
II
[¶ 4] On appeal, Moms argues the district court erred in awarding Moller primary residential responsibility because sufficient evidence was presented to establish Moller used methamphetamine while pregnant with C.M. and J.M., triggering a rebuttable presumption that Moller may not be awarded residential responsibility. Morris also claims several of the best interest factors should have been found to favor him. He finally contends the court erred by relying on the parenting investigator’s recommendations, alleging the parenting investigator was biased toward Mol-ler.
[¶ 5] A district court’s award of primary residential responsibility is a finding of fact, which will not be reversed on appeal unless it is clearly erroneous. Miller v. Mees, 2011 ND 166, ¶ 12, 802 N.W.2d 153. “A finding of fact is clearly erroneous if it is induced by an erroneous view of the law, if no evidence exists to support it, or, although there is some evidence to support it, on the entire record, we are left with a definite and firm conviction a mistake has been made.” Doll v. Doll, 2011 ND 24, ¶ 6, 794 N.W.2d 425. When applying the clearly erroneous standard of review, we do not reweigh evidence, reassess witness credibility, retry a custody case, or substitute our judgment for the district court’s decision merely because we may have reached a different result. Miller, at ¶ 12.
[¶ 6] District courts must award primary residential responsibility of children to the party who will best promote the children’s best interests and welfare. Miller, 2011 ND 166, ¶ 11, 802 N.W.2d 153. A district court has broad discretion in awarding primary residential responsibility, but the court must consider all of the relevant factors under N.D.C.C. § 14-09-06.2(1). Doll, 2011 ND 24, ¶ 8, 794 N.W.2d *269425. The district court also has discretion in deciding what weight to assign to a parenting investigator’s conclusion, and the court is not required to follow the parenting investigator’s recommendation. Id.
A
[¶ 7] Morris first points to factor j of the best interests factors, which states, in relevant part:
If the court finds credible evidence that domestic violence has occurred, and there exists one incident of domestic violence which resulted in serious bodily injury or involved the use of a dangerous weapon or there exists a pattern of domestic violence within a reasonable time proximate to the proceeding, this combination creates a rebuttable presumption that a parent who has perpetrated domestic violence may not be awarded residential responsibility for the child.
N.D.C.C. § 14-09-06.2(1)(j). “Domestic violence,” as used in factor j, is defined as “physical harm, bodily injury, sexual activity compelled by physical force, assault, or the infliction of fear of imminent physical harm, bodily injury, sexual activity compelled by physical force, or assault, not committed in self-defense, on the complaining family or household members.” N.D.C.C. § 14-07.1-01(2). Morris asks us to hold, as a matter of law, “that meth use by a mother during pregnancy constitutes domestic violence involving the use of a dangerous weapon, and any repeated use during pregnancy constitutes a pattern of domestic violenee[.]” He argues “the credible evidence did show repeated use of meth during the pregnancy of both children; such evidence constitutes a pattern of domestic violence directed at the child and creates a rebuttable presumption that a parent who has perpetrated domestic violence may not be awarded residential responsibility for the child.” Moller replies that, in the proceedings below, Morris “never asserted [Moller’s] alleged drug use while pregnant was domestic violence triggering the rebuttable presumption,” and claims the argument is therefore waived. Moller also notes the district court did not find sufficient evidence of domestic violence to affect its determination of parental rights and responsibilities.
[¶ 8] It is well established that arguments not raised before the district court cannot be raised for the first time on appeal. See, e.g., S.H.B. v. T.A.H., 2010 ND 149, ¶ 12, 786 N.W.2d 706. In the proceedings before the district court, Morris alleged Moller used drugs while pregnant with C.M. and J.M., and Morris voiced his concern that such drug use may have caused developmental delays in C.M. Yet prior to this appeal, Morris did not assert that a mother’s use of drugs during pregnancy constitutes domestic violence under factor j. Morris never specifically raised this issue in any of his district court filings, nor did he argue this issue before the district court. At trial, Morris addressed several of the best interests factors in closing arguments, but he did not discuss factor j. When discussing other best interests factors, Morris contended Moller was “denying ... and lying” about her alleged drug use during pregnancy, and he claimed “[h]er credibility ... should be very low under the circumstances!.]” Morris never directed the court to consider whether “domestic violence” as used in N.D.C.C. § 14-09-06.2(l)(j) encompasses the use of drugs by a mother during pregnancy, and the court did not make a finding regarding whether Moller used drugs while she was pregnant. With regard to factor j, the district court found, “Although there is testimony of arguments and name calling by the parties, there is not sufficient evidence of domestic *270■violence to affect the determination of parental rights and responsibilities in this case.” The court’s finding reflects that it was not asked to consider the argument Morris now raises on appeal. “If a party fails to properly raise an issue or argument before the trial court, the party is precluded from raising that issue or argument on appeal.” Id. Because Morris did not properly raise the issue of whether drug use by a pregnant woman constitutes domestic violence under N.D.C.C. § 14-09-06.2(1)0') and it was not considered by the trial court under the facts of this case, we cannot consider the issue on appeal.
B
[¶ 9] Morris next argues the district court erred by not finding factors c, e, f, and j favored him. Section 14-09-06.2(1), N.D.C.C., lists those factors as:
c. The child’s developmental needs and the ability of each parent to meet those needs, both in the present and in the future.
e. The willingness and ability of each parent to facilitate and encourage a close and continuing relationship between the other parent and the child.
f. The moral fitness of the parents, as that fitness impacts the child.
j. Evidence of domestic violence.
[¶ 10] With regard to factor c, the court discussed C.M.’s developmental delays, noting his pediatrician’s testimony that C.M. “is progressing nicely with therapy and should out-grow his delays.” While the court recognized Morris generally takes C.M. to therapy, it noted “[Mol-ler] testified that she does not go to therapy because of how [Morris] treats her in front of the children and that she does not want to interfere with the therapy.” The court then concluded both parties are able to meet the developmental needs of the children. We have stated that district courts must state factual findings “with sufficient specificity to enable a reviewing court to understand the factual basis for the court’s decision.” Doll, 2011 ND 24, ¶ 8, 794 N.W.2d 425. The court’s finding on factor c is not clearly erroneous because the finding is sufficiently specific and supported by the record.
[¶ 11] As to factor e, the court noted the parties are unable to communicate effectively with one another. The court found “[Morris] has told the children that he does not like [Moller] and he called the police in the presence of the children and sought to have [Moller] arrested.” The court also found Morris made inappropriate comments to the children about the parental rights and responsibilities proceedings. Morris asserts “the record is replete with [Moller] refusing to expand visitation, hindering visitation, and misusing the process against [Morris].” Morris does not cite to the record to support this allegation, and Moller and Morris’s mother both testified that Moller had fully complied with the court-ordered visitation schedule. Although discussed under another best interests factor, the court also observed, “[Morris’s] dislike of [Moller] was obvious from his testimony at trial and from the information contained in the exhibits. Under these circumstances, the Court does not believe that a joint parenting arrangement would work or that such an arrangement would be in the best interests of the children.” Evidence in the record supports this finding, and we conclude the district court did not clearly err in determining factor e favors Moller.
[¶ 12] Regarding factor f, the court simply stated, “Both parents appear to be morally fit.” Morris contends “[Moller’s] lying about [drug use during *271pregnancy] demonstrates a lack of moral fitness, which should have resulted in [Morris] being favored as to this factor.” Moller’s pregnancy records were admitted into evidence at trial. The children’s pediatrician also testified, noting he had never suspected drug use by Moller during his treatment of the children. Other -witnesses conceded they had observed Moller using drugs in the past, but none could say whether she had used drugs while she was pregnant. Under the clearly erroneous standard of review, we do not reweigh evidence or reassess witness credibility. Miller, 2011 ND 166, ¶ 12, 802 N.W.2d 158. We conclude the district court did not clearly err in finding both parents are morally fit.
[¶ 13] In addressing factor j, the court stated “testimony of arguments and name calling” had been presented, but the court found the evidence was insufficient to affect the determination of parental rights and responsibilities. Arguing the court erred in this finding, Morris relies only on his allegation of Moller’s “use of meth while pregnant and her continued lying about that use.” As we have already determined, Morris did not properly raise before the district court his argument that drug use by a pregnant woman constitutes domestic violence under factor j. Morris identifies no other evidence to suggest the court erred in not finding sufficient evidence of domestic violence, and we determine the court’s finding is not clearly erroneous.
C
[¶ 14] Morris finally claims the parenting investigator “showed a clear bias against [Morris] and as such her report and recommendation should have been disregarded by the Court.” Rule 8.6, N.D.R.Ct., applies to parenting investigators, and subsection (f)(3)(B) incorporates the Code of Conduct to which parenting investigators must adhere. The Code of Conduct “establishes principles of professional conduct and ethics aimed to ensure the proper performance of custody investigators in North Dakota.” N.D.R.Ct. 8.6, Appendix. General principles are also outlined in the Code of Conduct, including the principles of independence and objectivity:
Independence
A custody investigator shall preserve professional independence in the discharge of the investigator’s duties. An investigator should act in accordance with the law, free from all other influence, rendering investigative services based upon the investigator’s best knowledge. An investigator should avoid any impairment of independence and must not permit professional standards to be compromised by external pressure.
[[Image here]]
Objectivity
When providing investigative services, a custody investigator shall adhere to the facts. Facts should be presented in as neutral and clear a manner as possible. The facts should speak for themselves, and conclusions and analysis should not involve exaggeration, adjectival assistance, or other editorial overemphasis.
Id.
[¶ 15] Morris alleges the parenting investigator “demonstrated a clear bias against [Morris] and interacted exclusively with [Moller] during the trial. [The parenting investigator] even went so far as to congratulate [Moller] on doing a great job as she left the stand.” At trial, Morris questioned the parenting investigator regarding her allegedly biased conduct:
Q ... I’ve noticed every — almost—at least five or six times you’ve gone *272over and talked to [Moller] and said, “Hi”; isn’t that true?
A I was talking to — I don’t even know her name.
Q You made comments. Matter of fact, today at 10:38 didn’t — when [Moller] got off of the stand — didn’t you say, “Good job,” and give her a thumbs up?
A I might have said, “Good job.” I don’t remember — I don’t know.
Q Do you think that’s appropriate to do that in this setting?
A I think she can use some encouragement. Why not?
Q Have you provided Mr. Morris any encouragement through this process?
A I haven’t — I haven’t had a chance to talk to him.
Moller denies the parenting investigator was biased toward her and argues the parenting investigator’s statement of “Good job[ ]” does not demonstrate “clear bias.”
[¶ 16] We emphasize the importance of parenting investigators strictly adhering to the Code of Conduct when rendering their services — at all stages of their involvement with a case. Speaking only to one party at a hearing and providing words or gestures of encouragement to that party are inappropriate actions at trial when taken by an individual charged with rendering independent, objective insight and suggestions. While the parenting investigator’s conduct in this case is concerning, we cannot say that the court erred in awarding primary residential responsibility to Moller merely because the award parallels the parenting investigator’s recommendation.
[¶ 17] We have stated district courts may not delegate their decision-making authority by adopting a parenting investigator’s report and recommendation. See Sailer v. Sailer, 2009 ND 73, ¶ 34, 764 N.W.2d 445. Although the court in this case ultimately agreed with the parenting investigator’s recommendation that Moller be awarded primary residential responsibility, the court’s findings on the individual best interests factors varied from those of the parenting investigator, indicating the court conducted its own analysis in arriving at its conclusion. “The fact that the trial court agreed with the custody investigator’s analysis and conclusions is not in itself clearly erroneous.” Id. at ¶ 35. A district court has broad discretion in determining the weight to assign to a parenting investigator’s report and recommendation. Miller, 2011 ND 166, ¶ 15, 802 N.W.2d 153. “[A] district court’s opportunity to observe witnesses and determine their credibility should be given great deference.” Sailer, at ¶ 35 (quoting Hanson v. Hanson, 2005 ND 82, ¶ 11, 695 N.W.2d 205). Here, Morris was able to bring to the court’s attention what he perceived to be biased conduct of the parenting investigator. The parenting investigator also testified that, in preparing her report and recommendation, she had reviewed questionnaires sent to people listed as references by the parties, interviewed both parties alone and with their children, and examined depositions and medical records. When she testified, the court was able to observe the parenting investigator and determine her credibility. We conclude the court conducted its own analysis of the best interests factors, the court’s findings are supported by the record, and the court did not clearly err by agreeing with the parenting investigator’s conclusion.
Ill
[¶ 18] Because Morris has not established the district court’s findings are clearly erroneous, we affirm the judgment *273awarding Moller primary residential responsibility of the parties’ children.
[¶ 19] GERALD W. VANDE WALLE, C.J, MARY MUEHLEN MARING and DANIEL J. CROTHERS, JJ., concur. DALE V. SANDSTROM, J., concurs in the result.